The original thread was on the "Roman" - viewtopic.php?t=19354
Sorry for picking on Richard's quote. A lot of people could have made it, Richard was unlucky in just being one of the most recent.
Actually, the results show that "armies with pikes" have generally done better than Roman armies. These Roman armies have little else than legionnaires, while many "pike armies" have much better supporting troops than the Romans. That "armies with pikes better than Romans" is a vastly different statement than "pikes generally do better than Romans". The game results do not show the contribution of "pikes" to victory compared to the other arms in those armies. Inferring that “pikes” are the main reason for the rankings is thus a statement of faith.RichardThompson wrote:If you want to look at statistics then go to:
http://www.slithdata.net/files/fog/rankings.html
Choose:
Rankings -> army rankings,
ELO or Points Per Game (as you prefer)
Click here for more options -> Armies with 50 or more games (for statistical significance)
The stats will confirm your suspicion that Pikes generally do better than Romans.
Honestly, I haven't played enough games to have an opinion on most of the alleged short comings of FoG on the various posts for FoG 2.0. But so far, I can see that skirmishers can cause a problem – but perhaps that’s not unrealistic problem and one that can be perhaps fixed by victory conditions or if you play, as I generally do, in a campaign setting by using objectives. I have had trouble with “supporting lines”, which I can work around if I don’t mind a “line of columns” for the support line. I don’t mind but can see that some people might.
I’ve had a fascination with the apparent avalanche of flaws to FoG that need fixing. I’ve adopted a conservative attitude. FoG has rekindled my interest in ancient/medieval miniatures and I’ve played more games in the last year than the preceding six. So I have to be convinced that there’s a problem, but without having played a lot of games I look for the logic of the various arguments. I can tell you that a statement that begins…
“The stats will confirm your suspicion…”
…is an immediate “red flag”; and, whether the statement is really true or not, it makes me wonder if it isn’t a case of confirmation bias. This is common approach in politics, legal cases and policy analysis…you start with a position and look for confirming evidence in order to sway someone’s opinion. That’s okay but, like anything in politics, one ends up judging the case more by where the person is coming from than by what they are saying. My background is scientific and therefore I look to the data, try to understand the problem and then, slowly….slowly, develop a theory and finally to do my best to tear the theory apart. I tell the young scientists that join my team that you should be so familiar with the flaws in your analysis that no one – no one should surprise you with their criticism. However, I admit that such rigour is not often seen.
I’ve really enjoyed the FoG database and have spent a good deal of time looking over the games results. It’s fascinating and I think there are things to learn in the database but not what I see people doing….which is to quote differences without a clue as to whether or not those differences are significant or random and, if significant, why they are so. There are a lot of factors to control, before one can significance and cause. These factors include controlling for player influences and time variances. I should point out that most statistical methods for measure significance are based on the normal distribution, but if it’s not a normal distribution you could have great difficulty in assessing “significance”. As well when a distribution looks like a normal distribution but has parameters (mean and variance) that vary in time you’ve got one really, nasty distribution to work with - and there are clearly time varying influences in the FoG results.
It’s speculative, but I wonder if the performance of some armies actually resembles the economic case where a new market is discovered. There’s not much activity, but then the market takes off with the first movers doing well. Then more and more companies are attracted to the market, saturating it and results are diluted. Finally something else comes along and the market declines. It’s just a speculative analogy…but it could explain results as much as some of the arguments posed by people with an a priori position to prove.
Back to “pikes” being better than “Romans”…
If the FoG database does indeed show that, then advocates of that position must then explain why…
…the Early Carthaginian Army (with only 1 average, armoured cavalry BG, only 1 armoured offensive spear BG and hordes of protected foot) is doing better (10.42 versus 9 points per game) than the Late Carthaginian Army?
….the Ancient British do better than Mid-, Late-Republican Roman or Principate Roman. Can it be all of those light chariots? Wasn’t there a case about chariots not being cheap enough or good enough?
…Hellenistic Greek do better than any other Greek, Macedonian or Successor army?
…100 Years War English is better than War of the Roses English?
By the way, the Graeco-Bactrians are just under the Principate Romans (the worst of the Roman armies), yet the Graeco-Bactrians have armoured, superior lancers, cataphracts, pikes, thureophoroi, lots of bow armed LH (with an option for a Saka ally). Yet there they sit….low on the list.
What I’ve actually like using the FoG database for is to look at the detail…a slightly –ever so slightly - more empirical approach. For example, the Christian Nubian army is doing well, very well of late….so I looked at which armies have routed these infidels. Here’s the list…
Hellenistic Greek,
Late Republican Roman,
Palmyrian
Early Byzantine,
Carolingian Frankish,
Nikephorian Byzantine,
Early Hungarian,
Ghaznavid,
100 YW English (Continental) – twice,
Catalan Company,
War of the Roses English,
Santa Hermandad Nueva Castilian, and
Aztec – twice.
Don’t get too excited about the Nikephorian army. They’ve also been routed three times by the Christian Nubians. The fascinating ones on the list, for me, are the Carolingian Frankish (isn’t that from “Wolves from the Sea”!!!) and the Aztecs.
Just a few things to think about as we flame-on about what’s wrong or not wrong and how to fix what’s broken or what’s not broken….depending on your politics.