Hacker News new | comments | show | ask | jobs | submit login
Ranking Some Generals in the History of Warfare (medium.com)
68 points by lukateake 10 months ago | hide | past | web | favorite | 39 comments

"Ranking Almost Every General in the History of Warfare"

Not even close to "almost every general". Even taking into consideration his own admission that he left out the Mongols, his analysis is obviously very, very Western centric. Somehow the omission of the most populous continent in the world and probably throughout history didn't dissuade the author from making such a bold claim. The most obvious omissions were the Chinese and Japanese generals. It's not as if those two countries didn't have a lot of wars, wrote a lot about war ("Art of War" anyone), or lack a historical tradition. Let's not forget the Indian sub-continent has its own history of warfare.

His comment about Rommel was kind of awful as well. Rather than trying to correct for why his model didn't generate some expected results, he tried to convince people that their perspective is wrong. Rommel may not be the impressive general he's popularly believed to have been but the author's understanding of Rommel is quite shallow. Rommmel's exploits goes back all the way to WWI. He was the youngest recipient of the "Pour Le Merite" when he captured over 9000 prisoners with just 150 soldiers.

This is an example of someone with technical skills applying those skills to a field they don't really understand and wrapping it up with a bold claim.

Garbage in, garbage out. If your data set comes from scraping wikipedia, you're going to have these kinds of flaws and omissions. What's the alternative though, other than to hire an army of grad students? If flawed, at least it's interesting. I wish the author had gone into more detail about counterintuitive results (like Rommel), but part of the point of an exercise like this is to find instances where the model disagrees with common wisdom. If you jump straight to rejecting the model without asking why, you don't learn anything.

Some other thoughts:

- Analysis is limited to results of individual battles. That's a very narrow slice of a general's actual job.

- He ties WAR to overall W/L, which isn't great but the data doesn't give you many options.

- The model rewards "underdog" wins. This sounds like a decent proxy for skill, but it seems like a big part of the job is avoiding being an underdog in the first place.

- Army size and casualty figures for anything pre-17th century (and that's generous) are extremely suspect.

Agreed. Trying to compare generals from ~10 different centuries without taking into account the specifics of each campaign, and context of those campaigns renders the results useless.

I wouldn't trust this model to compare Montgomery to Patton during the Sicily invasion, let alone compare Caesar to Napoleon.

One thing I would be curious at generating from this data set, given the historical period spanning, would be correlation between expected battle outcome & actual battle outcome.

As someone who picked up a computational military modeling course in college, attempting to model ancient warfar is a vastly different task than modern battles.

My gut would be that modern warfare de-correlates more strongly from numeric advantage due to increased speed and lethality of available force types.

Also, for the author, if you wanted to be more accurate, start calculating actual expected outcomes from the forces. Lanchester's Laws are as good a place as any to start.


Eh, scraping any encyclopedia would introduce insurmountable bias and error. As you say, only primary sources are usable here.

Yeah, didn't mean that as a knock against wikipedia specifically.

>This is an example of someone with technical skills applying those skills to a field they don't really understand and wrapping it up with a bold claim.

Except he forgot to close with "which is why you should fund my startup."

OK, we'll put "some" above.

Has HN considered color coding title changes, or even the portions of the title changed in smaller or subtle changes like here?

Can't say we have. What would be the benefit?

1) People reading comments where people complain about the title get an indication that it was changed, so there's less "The title looks fine to me", "oh, they changed it already" comments that do't add to the discussion in any way.

2) If only the changed portion of the title is colored, there's an indication of whether it was a subtle changing to better fit the article content (even if the title was the same as the article previously) or an entire rewrite because it wasn't descriptive enough or was just plain misleading or flame-bait.

3) It may help you recognize it from the main page if you see it later, knowing it previously had a different title.

4) This is something separate from what I asked about before, but if all comments from HN staff noting changes followed the format Changed the title from $OLD" to "$NEW" because $REASON I think that it would cut down on the fluff comments about whether it changed (as in #1) and also cut down on the comments that opine about HN's editorializing of titles which seems to occur fairly commonly when the title changes on submissions of sufficient activity.

#4 would probably help the most, but even an orange asterisk after the title in a span with a title tag with the original title would be good. I'm just against loss of information in general, and would like to know what the original submitted title was in some cases, and it's not always obvious. Maybe I'm just weird.

It's almost 2018 and you're still peddling intolerant anti-sensationalism wrongthink? Get with the times! /s

This is really neat. I love the idea of applying sabermetrics to historic generals.

However, one failure of this approach is that it assumes that the numbers of soldiers given for ancient battles is accurate. The Greeks will tell you that the Persians came with millions of soliders, but that just wasn't even possible given the state of military logistics at the time.

So, look at the #4 general in the article, Khalid ibn al-Walid. As an example data point, the article on the Battle of Yamama (https://en.wikipedia.org/wiki/Battle_of_Yamama) says he fought at nearly a 4:1 disadvantage, and as far as I can tell, the only source for these numbers is the Quran. Much like how the Greek sources love to tell you how big, bad and numerous the Persians that they defeated were, how great is God's glory if the army of God is steamrollering equal or smaller size armies?

I'm not sure how the author could account for this. IMO this underweights more recent and well-attributed generals such as Napoleon or Wellesley.

>No other general came close to Napoleon in total battles. While Napoleon commanded forces in 43 battles

Subutai had over 60 battles. He also conquered largest areas and coordinated simultaneous operations over massive distances.

Not to demean him, but Subutai also had massive technical advantages with his mongol horse archers. He pretty much did everything else well though, not just tactics but strategies including psychological intimidation, adoption of opponents technologies, and just plain smart strategic thinking.

There are several issues with this analysis. Perhaps a more modern example would serve to illustrate this. Gregg Popovich is probably one of the greatest NBA coaches ever. However, if you look at the talent on his team compared to his opponents, maybe his wins with that talent isn't so impressive. However, an NBA coach's responsibilities include acquiring and developing talent. So the fact that he had more talent than his opponent in a game is a plus not a minus.

A general is also responsible for the logistics of maintaining his army and also of choosing battles where they have the most advantage. Thus, if a general over the course of a campaign is always engaging in battles with enemies where he is at a terrain or numerical advantage, that is positive for the general, not a negative. If a general is constantly fighting at a disadvantage in numbers or terrain, than can be a sign of weakness, not of strength - especially if they lose.

Thus, wins above replacement isn't a very good measure of generalship.

In addition, the data is incomplete. Napoleon fought 43 battles losing five. His biggest disaster, the Russian invasion was also his longest distance campaign. In contrast, here is the map of Alexander's empire. https://en.wikipedia.org/wiki/Alexander_the_Great#/media/Fil.... Believing that the 9 listed battles of Alexander represent every battle his army fought while conquering an area this vast is very naive. The truth is that over the 2000 years since Alexander, the accounts of battles got distilled and what we have is a greatest hits version of his battles. In contrast, Napoleon was in an era when it is more likely that we have a more complete list of his battles.

Finally, in comparing Napolean with some of the other generals, it is important to remember that ultimately Napoleon lost on the field of battle. When it really mattered, he failed. Alexander and Julius Caesar, Subatai, Genghis Khan, etc all won their battles - especially the ones that counted.

Napoleon may have been a great tactician on the field of battle, but being a great general is more than that.

Also, i think one also has to look at historical relevance and technology.

Fighting battles on the eastern front in WW2 is significantly different then warfare during ancient times or even the 18th century. For one, warfare is far more deterimental to the rest of society, and fighting effectively requires a lot more effort in the modern world compared to early times in terms of logistics, production, technology and strategy. Also, "warfare performance" is very much influenced by the enviroment one fights. How does one measure who is a better general when one, for instance, is severely undersupplied or in complete foreign terrain? there are simply too many variables to take into account to do suchs a study across suchs an incredibly large timeline.

WAR? haha I mean... I guess that would lead one to try and apply some analysis to generals.

This seems like a fun exercise in modeling data and doing some quick analysis. Obviously not for doing any worthwhile analysis or grading our existing general officer corps or putting any other weights in there.

For true analysis of tactical prowess you're looking at battle-to-battle analysis I'd imagine and metrics just don't go back that far. "most flanks covered", "most miles gained", "most efficiency per round of ammunition"... rather dark either way.

Well done for the purpose and nice post about how to gather / munge data!

Out of curiosity, anybody know what Schwarzkopf's stats would have been?

Not sure the intent of the author was to cover operational-level commanders. Even if Gulf War I did count, coalition forces outnumbered the Iraqis in sum total, so Schwarzkopf's WAR probably wouldn't be very good.

How about ranking some ambassadors and heads-of-state who prevented or avoided war? Less thrilling, I know ...

Yes, this analysis assumes the objective was to kill your enemy. Another view would be that generals that won battles with the fewest casualties on either side (or civilian) would be the most effective.

Otherwise, a general who was willing to nuke the other side might "win" a lot of battles, especially against a "replacement" who is morally opposed to civilian casualties.

You can also win the battle and lose the war. Are generals that overcommit to win a battle really good generals?

On this note, Otto von Bismark might be one of the most accomplished heads of state of all time, what he managed was pretty extraordinary, but at the same time he was swimming with the historical tide, so it somewhat magnifies his accomplishments "great man" style.

I wonder how true this is of many generals as well.

Anything that ranks Wellington[1] below Napoleon[2] needs some rethinking, IMHO.

[1] https://ethanarsht.github.io/military_rankings/Arthur%20Well...

[2] https://ethanarsht.github.io/military_rankings/Napoleon.html

And Wellington was a Field Marshal, not a General, in many of these battles.

So was Napoleon.

I think we are using "general" as a functional title here, not a rank.

One time, while playing Risk, two defending armies killed off 18 invading armies, while losing only one, and caused my opponent to immediately ragequit. That general must have been incredible.

The probability of accomplishing such a feat by pure chance is .00004 . But then the campaign was over, and that general never fought another battle, with a total WAR (according to the article) of 0.99996 .

Do you see my problem with the article? It rewards generals that fight unnecessary battles, completely discounts the decisiveness of the victory, and assigns a nearly arbitrary value for average replacement general. It is clear to me that the metric is measuring belligerence and battle prowess, rather than overall efficacy. The article even says it discounts strategy.

If, for instance, the general lured the rival's army away from the capital, while suffering a decisive defeat, but married his son to the widow empress during the battle, so the son could order the execution of the rival, the battle tactics were completely irrelevant in the face of the instantly-winning strategy.

A major problem with this analysis is not taking into account the number of battles fought. The article even states this:

"Alexander the Great, despite winning all 9 of his battles, accumulated fewer WAR largely because of his shorter and less prolific career."

In baseball, WAA is a measurement of wins above the average player and is actually a far better measurement for Hall of Fame worthiness. Even better is a variant, peak WAA, is a mathematical measure of how far above average players a player reached.

WAA helps you compare between those with long careers that weren't great players and those who were truly great players who had short careers. The author needs a WAA for this comparison, not WAR.

This is a solid idea, but his dataset is too messy and incomplete to draw conclusions with much certainty. He notes that it misses many ancient and especially Eastern generals just because they don't have Wikipedia entries, but even for the ones he does have the data isn't consistent enough to support his methodology IMHO. To do this right would be a lot more effort.

I imagine the average ability modern generals receive is due to far more accurate troop count measurements. That also explains the Western focus, as Wikipedia often doesn't have troop counts for historic battles elsewhere. Probably helps Caesar and Napoleon as well, both excelled at propaganda about their feats.

With a smaller focus, and the inclusion of some of the more obvious factors left out, this might have been interesting. In its current state it misses so many factors it renders itself less than worthless.

It's a ranking of a few popular wikipedia articles, based on metrics drawn up without any connection to the realities of war.

Was Borodino really a victory for Napoleon? Wasn't it more of a draw? Sure, the Russians retreated afterwards, so it didn't stop Napoleon's drive on Moscow. Still...

It was a Pyrrhic victory. At the end of battle, the Russian army retreated from the field and the French remained - so it was a victory by definition. But the French army could not replace its casualties or secure its supply lines, so the victory in battle resulted in defeat in the overall war.

I can't seem to comment anymore on the original article, so I'm going to write here in the hope that the author will see it.

Lots of people are commenting about the poor data set selection. I think that's understandable given that the author isn't a historian but rather a baseball geek and data scientist. Although I can't help but feel a little bit of anger to see (speaking as a software engineer and a history Ph.D. dropout) yet another example of a technical person blithely wandering into a field that has been studied for thousands of years and make very grandiose claims without even a cursory study of the field. Buy hey, that's tech for you, always disrupting (I mean that both sarcastically and not sarcastically at the same time).

What I want to address is more fundamental than getting the data set right, however. The author doesn't seem to understand that the very nature of the historical record is highly subjective.

1) Even if Wikipedia nailed every statistic, the statistics themselves about wins and losses, troop numbers, lengths of battle, places, etc. are increasingly unreliable as you go back in time. In some texts and historiographical traditions, the numbers are not just unreliable, they're arguably cut from whole cloth. Anything past, say, 1000 A.D. in Western Europe, for example, is highly disputed. Biases in old texts aside, we have enormous gaps in what texts have survived to this day. History isn't written by the victors, but the dried wood pulp and calf skins that it's written on is selectively preserved by them. The author's model has no awareness of the history of historiography.

(Tangentially, Napoleon, whom the author's model rates as the greatest general of all time, was indirectly responsible for a huge amount of destruction of Europe's archives after issuing orders to transfer archives from across the continent to Paris. Early modern logistics meant that huge portions of documents were destroyed in transport. I remember when I was working in the Vatican Secret Archives, something like 1/3rd of that archive was destroyed, and that's one of the main archives for European and world history.)

2) Even if we had all the numbers exactly perfect, what counts as a victory and what counts as a loss is highly subjective. Was the North Vietnamese Tet Offensive a victory or a loss? For whom? In what sense? These philosophical questions can't be answered by a model, at least not without the philosophical assumptions being made explicit.

This question is fundamentally one that requires a nuanced approach. I think that data-driven approaches can really help, but the author's model needs not only more refinement, it also needs to acknowledge more of the confounding factors involved. I encourage the author to keep working on it.

This is really cool, surprised at some of the output. Grant is really that high?

You should try this for admirals !

A major problem with ranking Mongol generals (and a lot of others) in this list is that there simply aren't historical accounts of most of their battles. Yes, the Mongols did send a large scouting force into Europe which wreaked havoc that entered recorded history in several places. That was a scouting force, not anything approaching what the Mongols would have considered one of their main armies. There are some accounts of travelers finding the aftermath of some of their battles in the near east, but much of their activity went on in places that did not keep such records. Just stitching together an empire in the steppes would have required a tremendous string of campaigns that not even Chinese historians would have any knowledge of.

Another major problem with this ranking is that, in many cases, the battlefield was not "level". Sometimes literally, but also technologically, socially, etc.. For example, Napoleon's Grande Armée was quite different from any other European force in its composition, supply, and leadership. It wasn't a small group of nobles carrying on knightly traditions, like most other European countries had. It was every freakin' Frenchman that could hold a rifle. For supply its soldiers relied heavily on foraging and raiding, which allowed them to move faster than opposing armies could. Command was highly distributed and Napoleon's commanders had a lot more information and leeway to act. When big battles happened, command chains had a way of breaking down, and that put traditional, strictly top-down command structures at a disadvantage. Napoleons army was bigger, faster, and smarter (if left to its own devices) than other European forces. It's arguable that Napoleon won so many battles, not because of his own tactical performance on the battlefield, but because he built a better army than those his peers built. Is that tactics? To put it another way, would Napoleon have fared as well against other French commanders had a schism occurred and he been faced with a civil war against half of his own army?

Julius Caesar had an even more overwhelming advantage over most of his opponents, thanks to the extreme disparity in organization between Roman legions and the Gauls or pretty much anyone else who wasn't Roman. He did spend some time fighting other Romans, but which type of opponent shot him up this list? Also, Caesar is probably one of the luckiest military commanders in history. If not for his luck, Caesar and his forces could have been annihilated in a dozen different battles and history would just have said "Well, that was a poor choice". How much does Caesar's luck inflate his position on this list?

Finally, one huge difference between baseball and historical battles is the historians. You can get a pretty unbiased view of any baseball game in the last few decades because you can actually watch a recording of it. Historical accounts of battles, on the other hand, are subject to all kinds of bias, exaggeration, etc.. Just the numbers of combatants involved become notoriously unreliable in all but very recent history, if they're even reliable then. For example, casualties inflicted by U.S. forces in the Vietnam war are often thought to be wildly inflated because they were often just assumed. e.g. U.S. forces take fire from somewhere in a jungle. They call in a napalm strike. Everything burns and the shooting stops. The U.S. commanders have incentives to pad their body counts, so they assume everybody shooting them (seemed like there were hundreds of them!) was wiped out. Accounts of battles just aren't trustworthy in a quantitative sense most of the time.

Almost every battle Julius Caesar ever fought was win or die. He was outnumbered by the Gauls in every battle, and the Gauls started adopting Roman organizational tactics after a while. And he didn't just fight a few battles with Romans, more like dozens.

At some point it's not just luck. He was a brilliant tactician and did innovative new things almost every battle.

The one battle Caesar did lose was an example of Caesars luck. Caesar only survived because Pompey (another great general) neglected (or was unable to) finish off Caesar's forces in the rout. Pompey still trapped Caesar's forces, who were sure to starve and dwindle if Pompey just waited them out. But his rich backers goaded Pompey into fighting Caesar again, a seemingly easy battle given his huge advantages in men, especially cavalry. But do or die Caesar pulled out another victory through innovative tactics, and that victory made him the ruler of Rome.

Yes, lets rank those who were instrumental in killing others, give us all something to aspire to.

This isn't a ranking of number of people killed. It's a ranking of how many battles were won given their resources. You don't necessarily win a battle by killing as many people as possible.

I think anyone who is remotely reasonable would agree that sometimes war is necessary, such as in wars of national survival like for the UK in the Second World War.

If we've agreed that war is sometimes necessary, we better make sure that we have people who are extremely good at it, which this is recognising and analysing.

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