Hacker News new | past | comments | ask | show | jobs | submit login
FanDuel not honoring bet that would have paid over $82,000 due to line error (espn.com)
163 points by ikeboy on Sept 19, 2018 | hide | past | web | favorite | 224 comments

Yeah, a quick read of the relevant regs tells me Fanduel is gonna lose this one (Unless i've missed something)[1,2]

The law on this is pretty clear, and they can't unilaterally decide not to apply the gaming regulations of the state, or change them through their contracts.

(They can decide not to do business in states they don't like the regulations of course)

BTW, somewhat hilariously, the part they "..."'d is just 4 words.

[1] http://state.nj.us/lps/ge/docs/Regulations/CHAPTER69N.pdf

[2] (c) Upon accepting a wager pursuant to this chapter, a wagering cashier shall cause the wagering system to generate a wagering ticket.

(IE them generating the ticket is acceptance)

(d) A wagering operator shall not unilaterally rescind any wager pursuant to this chapter without the prior written approval of the Division.

(h) Winning wagering tickets shall be redeemed by a wagering cashier after verifying the validity of the wagering ticket through the wagering system.

(IE they should have redeemed it because it was valid)

(k) No wagering ticket shall be voided after the start of the wagering event on which a wager has been placed

Another reason the could not have voided it.

Essentially the best argument they have is that it's a house rule under 13:69N2-2 . This only works if their internal controls were approved by the division. Even then, a reasonable court would likely find that clearly contrary to the written regulations (and thus, not a reasonable interpretation of the regulations by the commission).

I also wonder what they mean by mistake. I’m guessing these are set algorithmically; you can’t honestly label the times you lose money by trusting your code as mistakes while keeping money from the times it’s successful.

I wouldn't trust a surface reading of one section of the regulations, unless you're a lawyer with specific expertise in New Jersey gambling law and precedent.

There could easily be other offsetting law, regulation, or precedent. There might have been a prior decision which effectively gave "written approval" to void future egregious odds/payout errors. There may have been "prior written approval" of the house rules allowing recision.

Regulators & the industry have a converging interest in a predictable business, without "gotchas", and with precedents that protect against other potential larger future errors.

I would bet even-money that Mr. Price, the bettor, is never paid the full $82,610, either as a private settlement or at the direction of any regulator or court.

1. I am a lawyer, licensed in 4 states (among other bars) :)

2. I grew up in NJ and am quite familiar with the gambling regs (but i admit to not having paid attention in the past few years)

3. I checked with other NJ lawyers.

I am not a lawyer. I grew up in NJ but I'm more familiar with the policies of Nevada gambling firms.

But, I have a perhaps-unreasonable amount of confidence in my understanding of regulator-regulatee dynamics. I further think FanDuel competently understood their legal environment before making their statements defending the non-payment to the press. (If their experts knew they'd eventually have to pay anyway, why take the PR hit?) And finally, I believe that on general principles of fairness, such blatant errors should be voidable.

So, would you like to bet me $20, even-money, that your "FanDuel is gonna lose this one" prediction comes to pass? Me, a rando non-lawyer armchair commenter, versus your many credentials, familiarity-with-the-gambling-regs, and confirming opinions from other NJ lawyers?

You win if FanDuel ever pays this bettor the full $82,610 – whether due to public outcry or any action by relevant legal authorities. I win if the bettor accepts a settlement for less, such as the tokens-of-apology that FanDuel has already offered.

"But, I have a perhaps-unreasonable amount of confidence in my understanding of regulator-regulatee dynamics."

I also lived and worked in DC for 10 years :-)

I'm just not gonna bet or gamble here[1]. I expressed an opinion. I stand by that opinion. I will note you twisted it a little bit. I also gave what FanDuel's strongest argument will be, and I expect if they do win, it will be on those grounds.

But i do think you are very mistaken about at least one thing:

"I further think FanDuel competently understood their legal environment before making their statements defending the non-payment to the press. (If their experts knew they'd eventually have to pay anyway, why take the PR hit?)"

Because there is no meaningful PR hit - gamblers are not going to stop using it either way.

I also strongly doubt they "completely understood their legal environment" before making that statement, because pretty much no companies fanduel's size do so. You are suggesting a level of organized legal process that is uncommon. :)

[1] There are also too many paths where fanduel loses but the bettor does not get anything meaningful, or the question doesn't get decided because the person runs out of money or their lawyer sucks.

Fair enough, but I can't see how I "twisted it a little bit".

I don't see how any reasonable reader can interpret "Fanduel is gonna lose this one" as anything other than they pay the full ticket price.

There has absolutely been a PR hit, already, because most stories and lots of comments (like yours!) have implied FanDuel is in the wrong, and disliked for their behavior. (Even if it's a negligible effect among significant gamblers, it's not good for their brand among casual and new players. I had no idea they were operating actual betting windows, as opposed to online real-money "contests", so I've lowered my estimation of FanDuel from "novel online innovator" to "same old casino sportsbook".)

And further, if Fanduel were to confidently say they wouldn't pay, and then "lose" in a legal action as you predicted, there'd be deeper hits. First, a second round of stories that "Fanduel loses". Then, many more gamblers emboldened, in future similar situations, to insist on payment-of-errors rather than accepting the casino's "house rules".

And I believe definitionally, there are no "FanDuel loses but the bettor does not get anything meaningful" paths. If the bettor "doesn't get anything meaningful", for whatever reason – including that the rules are murky enough the bettor gives up, or the courts/regulators are stacked against such complaints – then those are precisely "FanDuel wins" paths. As I'm predicting, and willing to bet on, while you're predicting they "lose".

> And finally, I believe that on general principles of fairness, such blatant errors should be voidable.

If they’re voidable when the house loses, then they must be void ab initio and refunded when the house wins, otherwise it's a system of fraudulently inducing bets.

I'm surprised your “general principles of fairness” are so biased in favor of the party with more power and control.

> You win if FanDuel ever pays this bettor the full $82,610 – whether due to public outcry or any action by relevant legal authorities. I win if the bettor accepts a settlement for less, such as the tokens-of-apology that FanDuel has already offered.

That's a bad bet, because even if FanDuel legally owes the full amount and the trial outcome was somehow not in any doubt by the claimant, settlement for less because reimbursement of costs is not guaranteed to the victor in US civil process would still be a normal outcome.

Yes, if against all odds the house won a bet with similarly-erroneous odds, the betters would be entitled to a void by the same principle.

Typos, data-entry-mistakes, system glitches, software bugs which remain even after normal levels of care, data-corruption, etc shouldn't create liabilities of arbitrary magnitude between people when the error is manifestly obvious. Relative "power" has nothing to do with it. "Control" maybe a little, but here the bettor had the control necessary to strike during the short erroneous window.

> That's a bad bet

Not according to ~DannyBee's "Fanduel is gonna lose" and "law on this is pretty clear" and "checked with other NJ lawyers" confidence, I would think. If the law is so clear, the bettor could possibly get interest or punitive damages, as well!

But if you'd consider taking some fairer version of the bet, suggest stakes/odds/victory conditions, and I might be game! My core prediction is that Fanduel will only pay some token amount as a gesture-of-apology, at their own discretion, and nothing close-to the full amount, because the precedent is that important to them.

> Yes, if against all odds the house won a bet with similarly-erroneous odds

“Against all odds”? We're talking about a event that erroneously gets posted highly attractive odds for bettors, with no special features of the original odds or probabilities. You'd expect, over time, about half of the bets for such contracts to be won by each side.

Right, I should have said "this bet". But the same goes for any other bet with clearly erroneous odds.

However, bettors choose which offered odds to bet (while books accept whatever matches their 'board', within limits). And, good bettor practice is to check slips before leaving the window.

Thus a bettor can in practice exercise more discretion/control, in such a situation of odds-error-against-them, than (non-managerial) window clerks. They can simply not bet at bad odds, in a way that's harder for the bookie'srote systems, and they can usually even cancel a slip printed with surprisingly bad odds, on-the-spot. So errors-in-the-houses-favor will become actual live slips, needing exceptional voiding, very rarely.

I’ll talk you on your bet. Don’t back out like Fanduel!!!

OK, we have a bet! How can I ping you if I win? (My contact info is in my profile.)

(Still open to more action, on same or similar terms, from ~DannyBee, ~dragonewriter, or others.)

Good reply, I'm wondering they they will try to tie it to how slot machines handle errors ("All errors void play") which just payback the bet but don't payout.

In most states the regulations covering slot machines explicitly cover error. one of the most interesting parts is I can't find any part of the wagering regulations that cover error here. It's entirely possible they should but as far as I could tell they currently do not

Setting the odds incorrectly isn't considered to be an "error".

If your slot machine is supposed to pay out 1000:1 and you set it for 1:1000, sucks to be you, but it isn't an "error".

Actually: I apparently stand corrected. The situation generally gets investigated by the state board who works out the appropriate remedy.

Slot machines go through an absolutely grueling certification process. Interesting to see if FanDuel is willing to subject their software to it.

Seems to me like they'll get in trouble for not consulting the Division and getting prior written approval before rescinding the wager, but not necessarily that the Division will side with the gambler. I am familiar, at least in Oklahoma, where gambling operations have famously been allowed to skirt on mistakes.

But these are still contracts, subject to the basics of contract law. Contracts are not meant to be gotcha moments where one side leverages the mistakes of another. Look to the booking of flights. A small error here and there on price will be enforced. But a monumental error leveraged by people who would otherwise not fly will nullify the contract (ie hundreds of people suddenly booking first class travel London-Johannesburg because the website underpriced such tickets).

So in this case, where the remarkable odds were spotted by someone who knew it was a probable error, and would not have otherwise made the bet except for the error, I think a court will nullify.

What if it were billion-to-one odds? Should a company be destroyed because of a single numerical mistake? There is a line and I think this situation crosses it.

What if it were a 1% fudge above normal? That's frequent. Is it the liability of the pundit to check global markets? A billion to one odd would be unusual granted, but where is the cut off? Clearly this argument is not appropriate.

Conversely, bookies do not have to take into account whether pundits are likely making a mistake. Betting schemes thrive on the appearance that the bookie made a mistake. Either way, the places I know limit possible wins (e.g. 1000$ for a bigger, casual one). And even if not, they should be allow to take bets only if they can back it up.

That a pundit could not nullify the bet for the same reason is an edge the bank system abuses to no end. It's institutional. Since they don't have rigid rules for liability in computer security (e.g. in IOT and all the data ex-filtration cases) or code correctness only means the court would resort to opinionated decision on a case by case basis and precedent would only mean they can, not that they should, if the situation isn't exactly the same. There is bound to be precedent for exactly this case, which would open an avenue for abuse by the bookies. I don't think it's clear at all.

Well if you tell the Wall Street bankers... it’s not their fault and they require a bailout.

Yes they should honor it or else thy have 2 options.

1. Close business 2. Request a bailout.

Someone more competent should take their place. Fanduel is not above the law.

And yet, you can bet that they'd collect if the error were in their favor. And they probably have.

Absolutely disgusting.

That's because the terms (ignoring conflicting laws for a minute) choose FanDuel as the arbiter.

The imbalance is obvious even before the bet is placed.

I mildly disagree in the sense that what constitutes a "palpable" error is just too broad of a grey area.

It would be encouraging to see FanDuel just cough up the erroneous wager (which is surely cheaper than a lawsuit) and tighten down the definition of errors in spreads to something more specific such as anything outside like 2 standard deviations within the scope of a game*

*I'm not a gambler just someone who believes the definition of error can be mathematically defined for everyone's benefit

If it can be mathematically defined, then they can add that code to make that check in their system.

They would probably rather pay this out and have it go away instead of being a widely circulated ESPN article and potentially still losing in a lawsuit.

I think the reason they won't is they don't want to set a precedent for future similar mistakes. If they pay out here they would almost certainly lose future similar disputes in court.

I don't see why that should matter. What is the definition of a "palpable" error that the bettor can make and get his bet refunded after the fact? Of course none exists, it doesn't matter if your kid clicks a bet or you fat finger the wrong number or even if your own homebrew betting algo has a glitch, if you don't notice until after the bet has resolved you're just out of luck.

As long as there is no "I made a mistake, get out of jail free" card for bettors there shouldn't be one for the companies serving those bets either. FanDuel would have collected this man's money if the Broncos had lost and they would have collected his money if he had clicked this bet accidentally, to that end he deserves to collect his money.

Totally. Hopefully the judge orders they pay him over and above his deserved winnings if it gets that far.


What they could do is pay this guy off, then follow the good press for a very likely net gain.

Except then they have to pay out every other time there's a bug.

I imagine if they had to honor, they would make it priority number one resolving those bugs, and make changes to make the bugs more apparent to people in the process. If they don't have to pay, there isn't much of an incentive to prioritize the fixes, especially since I imagine if the bugs are in the favor of the company they can make a lot of money.

I'm sure it is already a high priority, but it is a system built by humans, and it will always have mistakes. What if they were being forced to pay out 82 million instead of thousand? What about 82 billion?

There should be clear cut laws surrounding clear errors, like there are in most places, and ones like this should not be forced to pay out.

When SCADA software fucks up and runs a ship aground someone has to pay. When something breaks because some engineer relied too much on FEAs and didn't do any destructive tests someone has to pay. "Whoops, our bad" is not an acceptable excuse except in tech.

"Whoops, our bad" is actually a valid excuse in virtually every industry, as long as your mistake doesn't cause damage that needs to be repaired. Such damage is the only reason anyone has to pay in the situations you described.

This situation is much more comparable to, say, ordering something from a store that turns out to be out of stock. The store has to refund you for the sale, of course, but that's all they are on the hook for. They aren't punished beyond that, they don't even have to go out and find the item for you.

There are legal arguments that become valid at those amounts.

There is a difference between:

Ouch, that stung! Must do better.

That amount kills the business!

They own my whole enterprise now?

This is a sting. Quality control tax.

They should pay up, suck it up, maximize good will, and prevent it from happening again.

Given this gaffe, it is even odds they have pocketed more than this amount from errors going the other way.

Then thy go out of business and someone more competent should take their place.

Otherwise there is no honor and incentive for them. Bookies have to live by their honor.

in this case the odds jumped from -600 to +75,000 ($12 profit to $82,000) What if they went to +75,000,000? Would they have to pay the guy $82,000,000? As long as the mistake was obvious, of course not.

Litigate it. That will be the answer.

it seems that errors are not uncommon and when they are "obvious" they typically void the contract. not sure why people are freaking out here. if you logged into your bank account and saw an extra zero that was caused by a temporary CSS bug, do you really think you are entitled to it?

If I remember my contract law, you're not 100% correct. Errors in the contract can be used as cause for severability or termination in some rare cases, but errors in execution are generally not. They're one of the reasons we have contracts in the first place. It is perfectly acceptable to sign a contract that in hindsight was one-sided. It's the pre-execution that requires mutual benefit. Otherwise every gambling wager could be challenged after the fact as mathematically the deal is biased to allow the book to arbitrage a wager!

maybe i should not have used the phrase "void"; in this case the contract includes a clause that says, "where a blatant or palpable error is made in... prices offered... bets may be settled at the correct price at the time at which the bet was placed". so more accurately, the incorrect price is voided

i think the issue here is that in some states, there is an authority that needs to determine whether the price was an error, and that it cannot be at the sole discretion of the sportsbook?

No risk, no reward.

Let me rephrase:

No honor, no bets.

I would not place a bet with them now.

Sounds like a great incentive to fix the bugs to me!

Disgusting? So if you see a sign on a car for sale that says $19,95 because a digit fell off, the dealer is obligated to sell it to you at that price? People and systems make mistakes, that doesn't mean it's "disgusting" that they protect themselves financially from those mistakes.

It's more like - dealer sold you the car for $19.95, then came after you and demanded the rest $19k because of the error.

Reminds me of the story from SparkFun's founder where a nightly currency conversion batch process failed and set all prices to $0. If you had ever bought anything from them before I believe they honoured the purchase and sent out the order for free. If you were a first timer I guess you were S.O.L.

Seems reasonable to keep good faith customers happy while rejecting those who wanted to pile on.

Once I went to recredit my public transportation badge in a shop mandated by the city to do the transaction and they mistakenly charged me something like 15€ instead of 150€ (I don't remember the exact figure but it was a 0 missing). Two days later I got my badge blacklisted (and earned myself a loud and obnoxious buzz from the badging machine in the bus).

I had to make a call and go to the transportation office to settle the matter. They told me the shop was likely to suffer repercussions after the incident.

If you bought it online you bet they would come after you.

I once saw a price for a car listed on a dealer's website that was much lower than the car was worth. Turns out when it was entered into the system, it was entered as the base model instead of the upgraded version with more options. The dealer offered to sell it to me at the lower price anyways if I put a deposit down.

Which is completely different than a digit falling off a window, right?

No, it isn't. It's still a mistake. A programmer could have dropped a digit just as easily as a person failed to properly tape a digit to a window.

When you go to buy that car where the digit fell off the price, the sales person will tell you that the erroneous price is wrong, and tell you the correct one.

If for some reason the sales person does not do this and goes through with the sale at the erroneous price, sucks to be them. They had their chance to correct it and they failed to do so.

This is the same situation, except instead of a sales person, it’s a computer.

It seems to me that this is yet another instance where companies automate processes that previously involved humans, then compensate for the newly introduced errors by blaming the customer.

If you don’t want software bugs resulting in sales at absurd prices, maybe you should have a human review each order to make sure the numbers are sensible.

> this is yet another instance where companies automate processes that previously involved humans, then compensate for the newly introduced errors by blaming the customer

In either the digital or analog cases of your example, the end result is that the customer ends up paying the intended price. The only difference is that with the digital case, they might find out about it slightly later. How is this a "newly found error" in that case and how is it "blaming the customer"? It sounds to me like the potential for error has always been there and nothing about the assignment of blame has changed whatsoever between the two examples.

That’s not the end result at all. There are three possibilities:

1. The error is corrected before the sale and the customer pays the corrected price.

2. The error is corrected before the sale and the customer changes their mind.

3. The error is not corrected before the sale and the customer pays the erroneous price.

Companies are trying to change a 3 into a 1 by blaming computers and saying the error should have been clear to the customer, when in fact it’s their own fault for not noticing the error before completing the transaction.

To be fair, they may also be trying to change a 3 into a 2 (or providing that option), but "trying" is the key word. There are enough high-value and/or physically large items that aren't cars (e.g. HDTVs, furniture) where undoing the transaction can be a huge imposition, even not considering that the merchandise is now no longer brand new and the customer must duplicate effort in the purchase process.

The bet was already made, the details agreed upon. In your scenario if the dealer said "Yerp, whatever's on the window is whatever you're paying" and you hand over your $20 and they accept it as payment then yes, I would expect them to honor it.

Would FanDuel void your bet if you misread the odds?

Totally different issues. Making a bad sale and then later backing out, vs advertising a price that is not able to be confirmed at the time of sale resulting in no sale.

Lots of grocery stores give you the product for free up to a limit (say $10) if it is priced incorrectly or rings through different from advertised

That was a consumer protection issue from the introduction of scanners, to assure customers that the scanner wasn't a black box that would rip them off.

The price sticker is an advertisement, not a binding contract. See https://en.wikipedia.org/wiki/Invitation_to_treat

This is true in some countries, but not in others. Eg. in the Netherlands it is binding the moment a customer indicates they accept it.

(On the other hand -relevant to the topic- an agreement may be declared void in the case of "obvious error"; so Fanduel could possibly have been off the hook if they had been operating in .nl .)

A lot of places have consumer protection laws about price signage and labeling which take precedence over contract law.

I was under the impression that this was the case in America and Australia as well, but it's possible it is only true for the UK.

I know from first hand experience that it's true in NYC and in the Canadian province of Quebec. I think it's true in many more countries and many more places in the US and Canada, but as those two examples show, it's not always legislated at the top level of a county, even when there is a law or binding regulation about it.

Do you have any evidence to show this? Or are you just making an assumption based on your dislike for the gambling industry?

It's a nice reminder to stay away from the gambling industry. Any company that doesn't play by the same rules as you, doesn't deserve your business. Try voiding a bet you made due to your algorithm error once. I'm sure they would just laugh.

I had my account closed by a major betting site shortly after winning (and withdrawing) a few hundred euro, multiplying the original stake by around 10x in a series of bets.

These companies operate completely arbitrarily, and they really hate winners of course.

> Any company that doesn't play by the same rules as you, doesn't deserve your business.

Which companies do play by the same rules as you? I can't think of any time I've been able to compel a company to use an arbiter of my choosing for any disputes.

Online retailers will do the same thing. They misprice an item, you purchase it, then they realize the error and decline to ship and give you your money back.

Owned by Paddy Power. Not surprised they are welching on the bet, they make a great play on what good sports they are in their advertising but will ban you if you show signs of not being a complete mug.

I worked for a betting company for a bit, and they actually liked people who were great at betting. They get flagged along with the sport/league they are good at, and the traders used these people to adjust their lines.

If someone bets and wins often on a specific sport/league, and they make a big bet for it that doesn't really make sense, maybe the line is wrong or that person knows something everyone else doesn't.

That's why I'm very suspicious of Paddy Power's ownership of betfair. They won't ban people there because their angle is as a broker working earning commission but they do potentially gain insider knowledge that they can then apply in their conventional business.

"Insider knowledge" isn't illegal and they couldn't care less what the outcome is. They make money by attracting customers who aren't sophisticated. If they want to know what the "right" price is, they just look at the Asian books (they don't employ people to help predict outcomes either).

maybe you're right but it being a well documented fact that they discourage successful punters to the extent of circulating photographs to betting shops makes me wonder. If someone is doing unusually well on their exchange why wouldn't they use that data, PP are extremely good at maximising every edge they have.

It looks to be usual practice across the industry in the UK: https://www.theguardian.com/global/2015/aug/02/betting-horse...

That article mentions Barney Curley, I'd recommend his book giving a little back. In it he documents a coup he pulled off against the bookies on a horse running at a small English course Cartmel in the 70s iirc. The key point was that there was just one telephone at the course. Most bets are settled at SP which is the on-course starting price. If the big chains are getting a lot of money on a horse they'll get their agents to put money on that horse at the course to bring the price down and reduce their liability. On that day Curley's people put a fortune of money on before they bookies realised there was something up. They hid it by betting in doubles where there's less scrutiny but then they pulled the other horses out of races which had the effect of converting those bets into singles. The routine would be for the on-course agent to phone head office to see what they had to do. On that day, Curley had a burly crony hogging the phonebox as he tried to keep up with the health of his aunt who'd been taken ill. No one was able to shift him until the race went off.

I just looked it up and I think I was confusing two different events. Curley's coup was at bellewstown in Ireland while he wasn't involved with the Cartmel one.



That is the business model. How do you think they make money?If you want to bet in volume at better prices, use exchanges or an Asian book. But their business model is to spend a lot on marketing to attract unsophisticated customers and make it back on bad prices.

Our system has this exactly backwards. An individual should get the benefit of the doubt, and their mistakes should be mitigated. An individual human gets tired, hungry, emotional, etc and other such experiences which cloud judgement. A corporation on the other hand is supposed to systematically avoid these things.

Because of this, mistakes in the interaction between an individual and a corporation should favor the individual. However, it seems that the system favors the corporation.

Gaming companies are especially bad about this. Their whole setup is designed to take advantage of human weaknesses in evaluating risks. However, when they make a mistake they are often the first to say that the bet is invalid.

>>> industry practice, which specifically address such obvious pricing errors.

I am in two minds here. Firstly this is a "contract" - honour it.

Secondly, I understand the industry practise, but Indont think it is in line with society needs - that is a human bookmaker would never have given 750-1 with the team leading and minutes to go.

So either you write software that won't do that and take the risk, or you stop using software to rinse every last penny out of human addiction

my preference is that mistakes in the software can bankrupt you. That way you pay attention to the software process

Well put.

Finance has seen its fair share of this. As an example, I was ashamed to see the difference in reactions between when it was Knight Capital [0] vs Goldman Sachs [1].

For Knight Capital, the entire company floundered after a trading glitch, and they closed shortly after.

When it was Goldman who suffered a glitch, many of their trades "qualified for cancellation". Co-author of the Black Scholes formula even spoke up against this hypocrisy.

Both companies lobbied to the exchanges and the SEC, but one had more capital. If this is any indicator, a company like FanDuel may be let off.

[0] https://dealbook.nytimes.com/2012/08/02/knight-capital-says-... [1] https://www.ft.com/content/37fff9c6-0b36-11e3-bffc-00144feab...

Goldman have a strong political game. Even over those periods in which the Secretary of the Treasury is not one of theirs, they occupy lots of other appointed positions in government.

I couldn't agree more. Software that doesn't have these kinds of bugs is far from impossible, it just costs (possible an order of magnitude) more than gaming houses are willing to pay. The consequence of them not being willing to pay should fall squarely on their shoulders.

Exactly. The bookmakers are pretending that their software is human, and thus should have human fixes applied as if it was a real bookie. But they’re literally making billions of addiction, why should we have sympathy for this company?

I feel like we need the Second Software Crisis movement to address the bug riddled software were still making, it’s as though we’re just acclimated to bugs everywhere as being expected. Perhaps in 2100 they’ll look back on this time in software as we do on the early automotive industry today.

The problem isn't that we don't know how to write bug free software, it's that we choose not to because it's far more expensive. It's about business priorities, not lack of technical knowledge.

Did the early automotive industry move fast and break things? Almost certainly.

I think an important difference is how easy it is to deploy absolutely garbage software. It's just not as easy to make a million cars, or fix a million busted cars, as it is to distribute an app across a million devices, so there's less incentive to get it right when you can "move fast and break things".

Obviously from reading these comments, most people here are not gamblers and prob do not have much interaction with bookies.

With bookies your honor and reputation is #1.

When you gamble, you take a risk and the counterparts takes a risk. Then it is up to Lady Luck to decide.

If you take a bet, you can’t back out. Just don’t take the bet in the first place.

In this case the guy sure had his stars aligned. To not honor that is insulting to all gamblers who respect the unknown and risk taking. They broke the social contract between the gambler and bookie.

Fanduel absolutely does not deserve to be company and should be grouped together with all the shady overseas bookies.

This is why smart contracts aren't nearly as good as people make them out to be.

I disagree. In an ideal Dapp, pricing is determined by the open market or with a strong oracle - nothing novel here. If the contract was audited and deemed secure, then a pricing error shouldn't be possible. Payouts would be automatic and instant, as determined by the contract or set of contracts. Therefore, even if there was an error, the platform itself couldn't refuse to pay out a bet because they don't have a say in it. Can't be evil > Don't be evil.

Edit: to add, if users saw unresolved bugs in the platform, the contract could be forked and improved or users could just leave. If code is law, and that code is deployed publicly for everyone to see, then nothing is left to the discretion of "owners" because the "owners" don't have discretion to begin with.

I don't quite follow. Are you suggesting that because there can be errors or misunderstandings in all contracts, the non-smart variety is preferable since they allow for human intervention when there's a dispute?

Yes, that's the usual argument against "smart" contracts.

What do you mean? A smart contract would force them to fulfill the bet, no?

No, because as has happened before, the developers just go "oops, we made a mistake in the smart contract" and nullify the gains of anybody who profited off of it.

That was only possible because the miners have united. It's not going to be that easy in the future or on a larger chain (e.g. Bitcoin).

"non centralized block chains" and other hilarious jokes you can tell yourself

Yea that's the problem. Let's say the smart contract accidentally burns the money instead of paying it out due to a bug in the contract code. No recourse.

Yes, smart contracts give no recourse when there is an error.

Whether or not that is good here, in general it is an issue. That is a large reason why we have courts.

You can program space for mistakes into your smart contract, though (e.g. no payout without final approval). Everyone would know exactly what is going to happen when.

In that case what’s the point of the smart contract in the first place?

Explicit and final agreement about the whole deal.

like an actual legal contract?

It was said that court decides, legal contract is thus ambiguous.

But inserting a human final approval... courts will end up deciding.

Yes, but this time, it'd be explicit, including details - who decides, how quickly, etc.

You mean like a legal contract?

I'm genuinely unsure of what you're getting at here.

A legal contract is not explicit. There usually aren't any details about any court and its procedures, any dates regarding the court, nothing. A legal contract is implicit - relies on the legal system (that is complex and changes often). Making a truly explicit legal contract is way harder than it should be - and impossible in most countries.

How does a smart contract with a human stage fix that?

There can be varying degrees of human involvement. The smart contract could say "a human decides" - and a human decides. But the contract could describe bounds that the human has, limits on how much time he has, etc. The point of smart contract is that if it really is "a human decides", it's explicit - and I wouldn't enter such contract. Instead I'd ask the other party to be more specific and integrate better escrow service that has more specific contracts.

So you want smart contracts with human elements in them so you can reject them for better smart contracts with less human elements?

I don't want anything, I'm just saying what a smart contract is - a 100% explicit contract that has many possibilities including varying degrees of human involvement - all up to the contract itself. People have different tastes.

There is no consumer protection there. No method to deal with contracts that are already mistaken.

Moreover, the best way I can think of to deal with mistakes in a smart-contract is escrow in something like a 2 out of 3 multi-signature wallet. This essentially elevates the third-party handling escrow to an arbiter. It seems like the kind of thing we'd want to be run by actual courts. Yet, having smart-contracts with build in ability of courts to revert seems to defeat most of the purpose.

Surprised at most of the comments here. If the line error had been for $82,000,000 would you have expected them to honor it as well?

Whenever there's doubt side with the customer but this was such an obvious error that any reasonable better would have realized it. Had the better lost the money I'm pretty confident they would have refunded the bet as well (since the bets were voided).

Or 82 billion, or trillion... yes, I would expect them to go out of business. Same if they made a mistake that killed someone, for example. As others have said, what if the client (or they kid!) had made the bet online and entered a few extra zeroes (and lost) - would they have accepted "I made a mistake" as an excuse and canceled the bet?

You should not encourage a business to have a "head we win, tails you lose" policy.

But a client entering a couple extra zeroes is not a clean and obvious mistake. A person could reasonably bet either $5 or $500 for example.

On the other hand, offering something like a 1000x payout for an event that has a 50% chance of happening is a clear and obvious error.

If an error is egregious and obvious, it seems reasonable that a company (in this case FanDuel) not honor the deal.

Similarly, say a grocery store meant to have a "Buy 1 get 2 free" steak sale but accidentally keyed it in as a "Buy 1 get 200 free" sale. Nobody would expect the grocery store to actually give away 200 free steaks to every customer who bought a single steak.

> But a client entering a couple extra zeroes is not a clean and obvious mistake.

Neither is an oddsmaking posting odds out of line with your perception of thpe likelihood of events—if you are gambling, that's exactly what you are looking for normally.

> If an error is egregious and obvious, it seems reasonable that a company (in this case FanDuel) not honor the deal.

This bet was made over the counter: if there was such an obvious and egregious error, it should have been refused at the counter. Would FanDuel have tested the bet induced by the odds as null and returned the funds if the bettor had lost? If not, why should it get to nullify it when the bettor wins?

I am pretty sure that would fall under pricing error laws,and at least in California, they would have to honor it.

Yes. How is it different from the software error that led to $440 million loss for Knight Capital Group (https://en.wikipedia.org/wiki/Knight_Capital_Group#2012_stoc...)? I bet they would love to cancel the trades they executed and get a refund as well.

> Whenever there's doubt side with the customer but this was such an obvious error that any reasonable better would have realized it.

But, essentially every bet is exactly the bettor assessing that the odds represent an assessment of the likelihood of an event that is wrong; from the bettor’s position gross errors of the market (the other bettors whose actions are the input to odds-setting) that they should mercilessly exploit are indistinguishable from gross errors of the aggregation system which you suggest they should not.

Especially with a human-in-the-loop system for taking bets, the protection for the oddsmaker should be that their agent refuse to accept bets when there is such an “obvious(ly attributable to the odds-setting mechanism) error”.

This is pretty standard. I've had a really angry email from a bookmaker after betting on a line that was like 10% off. From the bookmakers POV if they had to honour these bets either a) spreads would be much or higher or b) spreads would be so high there would be no market for bookmakers. I would hope the bookmakers do the honourable thing and void all bets on the misplaced line, not just the losing side. It's pretty scummy to freeroll your customers.

In this case the punter must have known it was an error, so it doesn't seem moral to pick up the 82k.

Somehow everyone else seems to think otherwise, so maybe I'm just a sucker.

>In this case the punter must have known it was an error, so it doesn't seem moral to pick up the 82k.

If we lived in a world where companies did their best to be reasonable and fair to customers, particularly when mistakes occur, I would agree. However, considering most companies are quite hostile and willing to use any advantage they may have to maximize profits or minimize costs at the expense of customers I believe we need to act the same. If companies are going to hold us accountable and punish us for mistakes or errors to the fullest extent then we should do the same

That's a bunch of generalities, though; are the specific accusations about fanduel or records of them benefiting from mistakes like this?

"Well, they probably took advantage of people" isn't a very good argument.

I totally agree with you. I think a lot of the other criticism is a buildup of frustration against bad customer rights, but in this case I am aligned with FanDuel.

It is so obviously a genuine mistake that some dude is trying to capitalise on and make a big media hoo-hah out of. He's getting $500 and three free games, just take it and be happy.

>It is so obviously a genuine mistake that some dude is trying to capitalise on and make a big media hoo-hah out of

If the situation were reversed, and due to a genuine mistake FanDuel showed that the customer owed the company the money, do you think they would simply accept that it was a mistake and fix it? If I had to bet I would guess that they'd attempt to collect the money using any means necessary just like the customer

That’s not the debate though, that would be morally wrong obviously. The fan is trying to get money that he/she doesn’t really deserve, and it’s immoral.

Two wrongs don’t make a right, and setting a precedent that it’s a feee for all makes thing much worse for everyone.

That ship sailed ages ago.

Maybe, maybe not. And if that were the story we'd be having a different discussion. My points still stand though.

I don't think a company should have to shell out tens of thousands of dollars because a software bug resulted in a couple of extra zeros somewhere. Do you?

Quick edit: A thought experiment. What if the bug created odds orders of magnitude higher than this? Should FanDuel be forced into bankruptcy from the debt to the customer?

Yes. Fanduel should have to deal with the consequences of their results. If an individual accidentally had more of an illicit substance on them than was legally allowed, they go to jail. If you accidentally choke someone to death during consensual play, there are consequences. If Fanduel hasn't found it important enough to have checks in place for their odds, literally the most important part of their business, then there are consequences. The problem with today's society is that people in power are divorced from the consequences of their actions, while those without power tend to have the harshest version of the consequences thrown at them. How many times have copyright holders illegally made false claims with no action taken against them? The point stands if we don't have consequences for our actions, we're not a country of laws and all laws should be invalid.

Surely there's insurance for such errors?

Yeah this was my first thought too. My sincere hope is that the law's response is "Pay up, this is how your business works. If this kind of thing is going to be a problem for you, buy insurance for it."

If a small business was out 82k because of human error as a result of a faulty product (the algorithm in this case) that's exactly what the court would say.

My sincere hope is that you accidentally fat finger a button while banking one day and wipe out your checking account sending a transfer or something. We'll see if you feel the same way then.

There is a clear incentive for gambling operators to claim “error” anytime they might lose money.

That makes it somewhat different, no?

>you accidentally fat finger a button while banking one day and wipe out your checking account sending a transfer or something

Completely different situations

Though I see where you are coming from. These seem like two different cases. One is a contract the other is a transaction.

Oh a "line error". That explains it.

Wait - what? What the hell is a line error?

At Vegas sports books I have personally seen tickets invalidated due to error. It is not just FanDuel, this practice happens industry wide.

They will settle this for half.

It'll cost them that much to settle on top of the settlement. Lawyers aren't cheap!

This is pocket money for them. They should honour it and move on.

You might want to reconsider using “welching” in this context. And don’t change it to “gyping”, either.

We detached this subthread from https://news.ycombinator.com/item?id=18025244 and marked it off-topic.

I just now learnt the etymology for Welch. Thanks!

I don't know why this comment is receiving downvotes. Both of those terms are pretty derogatory.

Care to clarify?

They're old racist terms against Welsh and Romani people

Oh! I never knew that. But it seems obvious now that you point this out.

I've read the Welsh being described as loquacious, dissemblers, immoral liars, stunted, bigoted, dark, ugly, pugnacious little trolls, but I cannot confirm its truth or otherwise.

What a repugnant and confused comment.

Please don't post troll comments here.


The insults against the Welsh in that post are rather notorious being from a column in the Sunday Times by the late A.A. Gill. A complaint about them was thrown out by the Commission for Racial Equality. Not everyone, but many people in the UK view insults against the other nations of the UK as banter, a wind-up probably because we're all part of the same greater in-group.

I'm fine with welching, thanks, been using it since long before anyone suggested to me it might have origins such as you infer. In fact since you're the first person I've known to be offended by it, then unless you're Welsh, I suggest you take a break from virtue signalling, it's really boring. Never heard the term gyping until you mentioned it though.

Sure, back when I was a kid, both were in common usage, and I used them freely and unknowingly into adulthood. So, I was quite taken aback to belatedly learn of their etymology three decades ago.

I'm happy to hear your data point that "gyp" may already have fallen into disuse; or did you mean that you now intend to start using it now to signal your non-PC-ness? If so, others that I recall hearing in childhood but not lately, which you might also consider adopting, include: "indian-giver" (one who reneges), "jewing down" (bargaining excessively), "n....r work" (which seemed to mean "hanging around and not working" back in the day, but now Urban Dictionary has it as "easy, repetitive work").

One thing we can be sure of is that trying to direct the language people use is a sure fire way to provoke a reaction. That goes for your ridiculous request for me to not use the word welch in the sense that i used it. If it had an association with the Welsh people that's long since been lost and I say that as someone who's lived on the border with Wales.

I think most adults are fine with hearing, "hey that word comes from a bigoted place" and replying, "oh I had no idea, I'll try to use a different word in the future."

That sounds like a very odd conversation between two adults to me. Sounds more like a mother teaching her child. The word welch is fine, the fact that it has historical origins makes the word even more interesting. It's survived until now and we're no more enlightened than previous generations who also used it. In fact thinking we are, as we do, is one of the conceits every period seems to have.

There are many ethnic and racial slurs that were used in previous generations - that doesn't mean they are acceptable today. Surely you don't use some of the examples drfuchs gave even though they were once common place.

Of course I don't, we all conform, why is another question.

It seems like a bit of a stretch to assume the idioms someone uses can reliably indicate their moral character. As with anything, granting people the most charitable interpretation of their actions can go a long way. Is the person using the term gyping actually showing their hatred for the gypsies? It's not impossible, but that's not a very charitable interpretation of their actions

damn I looked at your profile!

This will just add to the (necessary - don't get me wrong) tax on society that is our legal system. It's pretty obvious what should happen here because, as humans, we understand and allow for the possibilities of mistakes.

Fan Duel should send him some tickets to a game, try to create a positive interaction for him, and let's move on.

What's obvious about that? To me the obvious thing is that bookies have to honor the bets they sell and get to eat their delicious mistakes.

It's within their power to improve their automated system to not offer stupid bets.

>It's within their power to improve their automated system to not offer stupid bets

So then it's also within your power to never write bugs. Have you ever written any? How did that happen? It was within your power not to do so, right?

I've never written a bug in a system designed to earn me money and then denied responsibility for the consequences of the bug, no.

That is, my argument is about how you decide responsibility, not about technical excellence.

So now it only matters in finance? No one lost anything here. It's ridiculous to expect a payout due to an error like this. You wouldn't get one at a casino if e.g. a slot machine malfunctioned. What if the payout was $500M? Should they have to pay that as well?

Actually you won't get a payout if the slot machine malfunctioned: https://vegasclick.com/games/slots/malfunctions

The difference being that the slot machine owners are not the slot machine developers and it is not within their power to do anything to prevent the errors.

thanks, that was a typo

Sure, I'm entirely comfortable making life difficult for bookies.

And why is that? The rest of society has deemed it a ligitimate business. So you simply don't like gambling and that's why you want them to eat it? Not a very strong argument.

It's more that I think the activity of promoting gambling is one of getting your customers to eat it (It's advertised as a game of stakes, it's really just skimming transaction fees), so I don't mind it if what goes around comes around once in a while.

Society did no such thing; they weaseled their way into a legal grey zone and made themselves big before society had much say in it.

And if that bug cost his company say, $80,000 for some reason, there's plenty of instances where his company just has to eat that.

I've literally made tens of thousands of dollars in mistakes over my careers (try not to make the same one twice) and 9 out of 10 times the company would go "shit dude, don't ever let that happen again!"

And what if it were 1M? 500M? 1B? They just go out of business? Your logic doesn't scale well and seems overly harsh. They offered $500 and a pair of tickets. The guy lost nothing, I think that's fair.

You may be surprised to learn that software bugs put companies out of business. They also can kill people.

I'm certain I wouldn't considering I've spent my entire career in cancer diagnostics and finance.

This is a completely reversible mistake and no one was harmed in any way (feelings don't count). Let's not go off the rails and start making Therac-25 like comparisons. This isn't some abstract problem; the issue is well defined, so let's stay within the realm of reason.

Do you also think bugs in HFT code should be able to "undo" their mistakes instead of losing lots of money?

And how do you propose that would be done? You've totally jumped the shark here. Comparing this, an easily reversible incident, to HFT is simply nonsense. Thousands (probably a lot more?) of transactions would have occurred within the context of the bug, and decisions were made based on bad data once the bug was in the wild. There's no way you could undo that.

If your company could create a ToS to get out of it, they would though.

There are reversible mistakes too.

It is very typical for contracts between businesses to include monetary penalties for bugs that are service or uptime impacting, so yes many other companies are required to pay for their major bugs, why not a betting company?

>that are service or uptime impacting

That's why. What's the impact here? The guy didn't lose anything; he got his hopes up and was disappointed, that's all. People here are acting like the degree of severity/impact doesn't matter, but of course it does.

What if there was another zero on there? How about 8 more zeroes? Still think they should need to honor the bet? They offered him $500 and tickets to a game, I think he is getting greedy.

What if the situation were reversed, and I bought a wager only due to the fat fingering of my calculator, or I mistook one team for another or just had a brain fart.

Wouldn't it then be reasonable for them to give a refund?

A part to consider is whether the other party knew or should have known of a mistake.

It's not reasonable for FanDuel to assume a mistake there. However, if a customer purchased a bet on a game that had already happened it would be reasonable to assume a mistake had been made.

Mistakes happen and the law of contracts deals with them.

Yes, and bookies do as long refund clients' mistakes as long as the price hasn't moved that much and it is before the event.

Both which were not true of this interaction.

What if there were less zeroes? At what point can the company just decline to pay anything out at all because it’s up to their discretion whether a line was good or caused by a bug? Obviously in that case they would go out of business because people would be mad. But my point is that (in my opinion) it’s a very gray area legally.

It was less zeroes and they didn't pay and the guy has threatened to sue. Obviously it's a gray area, which is exactly why there are 2 sides to this argument. Their fine print will either stand up in court, or it won't. But you can't just assume they are going to pay out willingly every time they make a mistake. There is obviously a line where they have to say the risk in not paying out is worth the bad publicity and lawsuit potential. Then they learn from that mistake and move on. Then people use your business moving forward or they don't. That's why I like free markets, the checks and balances are already there to handle situations like this, we don't need legislation.

It might be within their power, but they are going to need to collect even more juice to get the system to that point.

That or they might need to close up shop.

Most people prefer the current system where they are able to make very dynamic sports bets, sometimes in real time, with the rare mistake.

It's easy enough to imagine a gatekeeper that doesn't let high ratio bets through, or puts them in front of a human, or whatever.

Test the hell out of that and you are a long way towards where you want to be.

It kind of smells like they had one, except it only panicked after the bet sold...

> It's pretty obvious what should happen here because, as humans, we understand and allow for the possibilities of mistakes.

If a better makes a mistake, they still lose their money.

The same should apply on the other side.

They tried 500$ and 3 Giants tickets, which he declined in favor of getting a lawyer.

I think they should bite the bullet, pay it out, turn to their software engineers, and say "it's worth 82,000$ to this company to ensure that never happens again."

Its accidentally $82,000. If you allow such a bug to be honored, then potentially it could manage a near-infinite payout.

Such bugs becomes not a costly mistake but instead existentially threatening

And we certainly don’t have a software industry capable of supporting that environment

Hedge funds seem to automate things in an environment where certain bugs are existentially threatening. They hire doctors of math and get their code correct. It's not impossible, just expensive.

Right. If you make system creation more expensive, the house needs to collect more juice.

I think most sports betting consumers here prefer the current, lower cost, more diverse system with a slightly higher error rate.

I'd reckon most people complaining here don't even bet on sports, but want to grandstand against bookies. I can understand that I guess.

I'm not going to comment on moral or legal responsibilities of bookies and gambling companies, because I am uneducated on both subjects, but I found your use of the term "collect more juice" really interesting. It seems odd to relate pulling revenue from fruit to pulling revenue from people's incorrect bets. Is that an industry term?

This isn't a nonprofit charity. It's a bookie. They decided to offer automated betting systems, in the hopes of winning money from their customers. They should stand behind those same systems when they don't work in their favor.

If you can't afford the errors you should not be in the business (but they can easily afford this error).

I don't think so. The legal system in Europe honors good faith.

Luckily for the better New Jersey is not in Europe.

Oh yeah, it is - I just wanted to offer an alternative to your views, and added a real-world example.

We do too have software that can be that reliable. A aviation and space flight are quick examples. Gambling houses probably don't want to pay the cost of that software, but it could be done.

It wouldn't just be bookies paying that cost. It would be forwarded to the customers placing bets. I'd prefer to keep the costs low, since nobody is going to die and things aren't going to blow up. The worst that will happen is someone thinking they scored a huge payout and not getting it.

Sure, but in this case it’s not an infinite payout. And the courts will care about that more than an engineering hypothetical, i would wager.

The software industry absolutely does have ways to prove a much higher level of code correctness than Paddy Power is demonstrating.

Best use humans then. Otherwise there is no trust.

You know they did not inform people of errors the other way.

The best part is they did use humans, this bet was placed in person.

> The bettor, who identified himself to News 12 New Jersey as Anthony Prince, placed the wager over the counter at the sportsbook at the Meadowlands Racetrack

>Fan Duel should send him some tickets to a game, try to create a positive interaction for him, and let's move on.

They offered that and $500. He refused and got a lawyer.

> as humans, we understand and allow for the possibilities of mistakes.

Yes, but corporations aren't humans. If the mistake went the other way, it likely wouldn't matter. The human would have to deal with it. I say give him the $80k+ and learn to deal with your mistakes.

> Yes, but corporations aren't humans.

Mitt Romney gave your opponents their line: "Corporations are people, my friend"

PaddyPower's revenue is 1.75 billion. Why wouldn't Bill Gates just eat the cost of the mistake.

This isn't a "human interaction". A corporation isn't a human and is governed by the "make as much money as possible" theory (Shareholder Primacy), so why would we treat them the same as if this is a human interaction?

Eat the cost? Why doesn't Bill Gates just donate a few million dollars to every charity? Based on his net worth, that's not a lot. I think it's about setting the precedent.

I presume Mr. Prince's idea of a "positive interaction" is to get his $82k.

Do we really have much sympathy for a gambling company? If you make an error, they expect you to pay. Why shouldn’t they? Oh no, their huge profits take a one time $80k hit. I hope he wins in court.

Exactly, we shouldn't apply the law equally, we should apply it to how morally tasteful we find the business and how much money the company makes.

And what does the law say in this case? Can they refuse to pay out if they themselves decide it was a mistake to take the bet in the first place?

They did:

>FanDuel instead offered to pay him around $500 and give him tickets to three New York Giants games. Prince declined to take FanDuel's offer and told News 12 New Jersey that he planned to hire an attorney.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact