Double weight of current year stats
 

It sounds like they have fixed the problem of using Neutralized Stats in Historical Leagues....:)

So, I think I have my settings figured out for the most part in a league that I will be starting in the near future.

I will use:

• 3 year recalc with Neutralized Stats
• Base potential ratings on...remaining peak seasons
• Base fielding ratings on...3 year period
• Base pitching stamina on...3 year period

The one thing I am not sure about is using "double weight of current year stats." I know the theory behind it, but I don't know if it could have a major effect one way or the other. I am inclined not to use it.

Why would you use neutralized over real? I’m still figuring some of this out, why would you prefer that one?

Perhaps this?....It helps to remove some outliers.

So, they'll be rolled out in the next update? That would absopositively fantabulous! :happy:

I'm using them for testing purposes to see which I prefer. I'll probably do a 3 and 5-year double weighted recalc with Real Stats and with Neutered Stats and then pick one and go with it. Should take quite a while, but I'm quite sure the game will be being patched, not because it's bad, but because it's software, and software always needs patching.

I'll have players in the league who didn't have a lot of MLB stats, or no MLB stats, i.e. Negro League players, Japanese players, etc.

Neutralized stats will correct that disparity.

I'll draft those players from Spritze's DB.

Any opinions on using doubled weighted current year stats?

IÂ’d use it. If you are concerned about outliers, you can also enable the DH if you havenÂ’t done so already. Also turn off injuries.

Why would you turn off injuries? you might want to set them to low for a few seasons to build up a reserve roster and/or free agents but turning them off compounds the outlier problem because guys get too many AB's.

FWIW I am in the 5th year of a random debut that I started at the beginning of beta and use the classic injury setting and high fatigue for position players and am quite happy with it.

With high fatigue, what kind of games played numbers are you getting? I'm really considering using high fatigue myself. Garlon posted some settings a ways back for historical leagues and I'm thinking of seeing how they work in a random debut setting. High fatigue and 5 year recalc were two of the main settings.

I think AB's and PA are more telling because if a guy PH's in a game he still gets credit for a game played. So here are my top AB's so far as well as the actual from Baseball Reference for 1986 which is the year I locked stats at.

FWIW the highest PA number I found with a quick look was 739 which seems right on.

The high fatigue seems to effect catchers the most obviously so you won't see any catching 154 games which is what I use for a schedule.

I still haven't heard any pros or cons on using "double weight."

If checked, I guess a player has a better chance of matching the year's total that is doubled.

For example, on a three year recalc, his middle year's stats are more likely to be matched, and not the other two years:

1965 - 41 HR
1966 - 32 HR
1966 - 32 HR
1967 - 27 HR

He is more likely to hit around 32 HR, I would think?

I'm afraid I don't have a lot to offer on that. I don't use it. I go with a straight 3 year recalc and I leave the OOTP engine activated as well.

Let’s say you have 3 yr recalc double weighted for current year. Let’s say play X IRL had
1980—25HRs
1981—25HRs
1982—50HRs
1983—25HRs
1984—25HR

In OOTP if you take ou randomness and other factors, you would expect
1981—32HRs
1982—37.5HRs
1983—32 HRs

Over the 3 year period the total will be the same but the chances for the player to lead the league in 1982 will be greatly reduced. (IE-Maris won’t break the record).

What happens when injuries and or fatigue is turn on? This is just conjecture on my part since I obviously don’t know how Markus created his game engine.
At the beginning of the season the game engine is looks at how many hits that it needs to create. Let’s say it needs 20000 hits. The engine will look at all your settings (injury, DH, fatigue, etc.). When it looks a the injury setting, it determines 2000 hits ( I just randomly selected that number) will be lost due to injuries (mostly starters and your better players). Their replacement might get a 1000 of those hits back but the other 1000 will go to other players. Therefore, players like Hornsby (and all other players) who don’t get hurt will get a boost to their hit total. Some of the players that were injured will have batting averages at .400 due to randomness (and some will be lower than their RL averages). Therefore injury (and your other settings) will definitely influence the game engine as it tries to maintain league totals. I hope this makes since, I just had a couple drinks and I’m in a hurry. I could be completely wrong but in my limited sample that I have done, it seems correct.

Not saying you are wrong, but I would like to know how you arrived at those expected numbers?

3-year double weighted recalc will provide these settings. Let's say the year in question is a player's 1985 season. 25% of the weight will come from his 1984 RL stats, 50% of the weight will come from his 1985 RL stats, and 25% of the weight will come from his 1986 RL stats. For five-year double weighted recalc, the breakdown for the same player would be 1/6 from 1983 RL stats, 1/6 from 1984 RL stats, 1/3 from 1985 RL stats, 1/6 from 1986 RL stats, and 1/6 from 1987 RL stats. You'll notice in both examples the weight for 1985 RL stats (the current year) is double the weight of the other years.

When the weights are taken into account are the project stats for the season when calculated based on 162 games? So if the player misses games due to rest or injury then he will not achieve his calculated stats made before Opening Day plus or minus random variation?

Let’s say a player only had 100 ABs IRL for the season but had 10 HR. His power rating will probably be very high and will based on the real life season. 10 HR per 100 ABs may be outstanding or could be average for the league that season. Then you put it in an OOTP league and if the season is basically the same as his real life season, then you would expect 10 HRs for every 100 ABs He has. So if he is expected to have 600 ABs you can expect him to have 60 HRs. If you put him in an OOTP season where HR are hit at a rate of half his real life season (maybe the mid 60s) then you would expect his HR rate to be half or if he is expected to have 600 ABs, you can expect him to have 30 HRs.

I would think the game is more interested in the rates that a player hit a HR or double in rather than the raw numbers themselves. This makes your Adjust/Weaken settings important. Too high and you might cause some guys to be Adjusted/Weakened that don't really deserve to be. Too low and you could create beasts out of guys with very few AB/IP. I use 200/50 for AB, and 40/10 for IP. Always have because those are ideal for Neutralized Stats, which I had always used until this year. Gonna give Real Stats some test runs as well and pick out whichever one suits me. Missed seasons are a big reason for using 3 or 5 year recalc because there are definitely some numbers to work with regardless of whether the player was serving his country in wartime or injured or whatever the case may be IRL. Jeff1787 has definitely convinced me that if you're going to use players from outside the MLB database, you probably want Neutralized Stats, but there were issues in OOTP18 with some HoF type OF becoming first basemen. I love using players from outside the MLB database, so that might force my hand. We shall see.