Quote:
Originally Posted by SteveP
No ... or maybe. The people who say that you should sim a couple of seasons before you actually start playing probably believe that the LTM system has to triangulate in on the correct levels. I don't think that's true, because otherwise this technique wouldn't work so well for the first season.
OTOH, the automatic LTM calculation is always playing catch-up to some extent. It's working from stats recorded in one season and making adjustments to be used in the following season -- but in the meantime, your rosters will have changed (retirements, rookies, players coming back from injuries, etc.). Based on that, you might think the automated process needs more time to "get it right." You might also think you should use my technique every season (which could be done). My conclusion is that it makes more sense to get the first season right, and then just let the automatic LTM process do it's thing. The resulting stats are always going to be somewhat "wrong" no matter what.  |
Thanks SteveP, I suppose I could run a test of the next season using both methods and see which one works better. Your method for the first year resulted in the following rate stats and ratios for 1901 (I basically ignore counting stats because there are all kinds of reasons why they'll never be right, but the ratios and rate stats are the accuracy-seeking baseball simmer's best friend):
Real vs Sim:
BA: .272 vs .271 (.27145)
OBP: .327 vs .324 (but remember, game tracks SF, which drags down OBP, where RL did not; take out SF, OBP becomes .328 [.32758])
SLG: .360 vs .361 (.36054)
WHIP: 1.352 vs 1.354
H/9: 9.7 vs 9.6
HR/9: 0.2 vs 0.2
BB/9: 2.5 vs 2.6
K/9: 3.2 vs 3.1
K/BB: 1.28 vs 1.19
FPct: .943 vs .943
SB%: 55.0% (Garlon-Spritze land) vs 57.0%
R/G: 4.99 vs 4.83
ERA: 3.49 vs 3.28
R/G minus ERA: 1.50 vs 1.55
CG%: 86.2% vs 75.0%
Other settings: No coaches/scouts; No suspensions or injuries (avg total roster size is about 23); High Pos Player Fatigue; Player Development Disabled; Traditional Lineup Selection with 100% Weight on Ratings; Neutralized Park Factors and Weather; No Minor Leagues (yet); Auto Adjust League Strategy On; Retire Players According To History On; 3 year Recalc - No Double Weight; Neutralized Stats; Base Rookie Fielding Ratings/Pitcher Stamina On: Entire Career; Auto Adjust LTMs after each season: Off to collect LTMs you will use, On With Those LTMs inputted for your sim; Auto Import Player Creation Modifiers On.
Notes: I used 10 sims in collecting the LTMs that I used for this sim, then took the average, and used it. 2 categories stand out as being way off even with proper LTMage: 1) Balks: which happen so infrequently, that I just don't care and 2) Errors. The first time I simmed I came out with a .980 FPct, which is just wrong. RL Errors: 5327; First go around errors: 1929. 5327/1929 = 2.762. Then I multiplied each of the 9 Position Error Modifiers by 2.762, to come up with proper ones. Next time through: 5316 Errors: much better.
SteveP, your method speaks for itself. Thanks for sharing.