The ratings calculution changes from v25 to v26

[uruguru]

I don't understand why this has changed in 26. Here's a good example. Attached are the 1962 ratings for Rico Carty, set to 5-year averages with no doubling of the current year.

This means that this 1962 version of Rico Carty's ratings should include his remarkable 1964 rookie season with the MLB Milwaukee Braves when he batted .330 with 22 homers in 505 PA.

The OOTP 26 ratings (on the left side) make him unplayable but the OOTP 25 ratings will put him squarely on the MLB roster.

I understand that his 1964 season should only be 1/5th of his ratings, but I would like to understand why the weightings changed from 25 to 26.

If I change the ratings window from 5 years to 3 years and exclude his 1964 season, the differences between 26 and 25 are much smaller.

[Garlon]

I think what is happening here is that in 1962 Rico was in the minor leagues. Your 5yr recalc is from 1960-1964.

In 1963-1964 he had 457 major league AB. From 1960-1964 he had 1208 minor league AB and most of those were below the AA level. I think OOTP is basically taking those minor league stats and knocking him down based on the MLE of those levels.

I created a 5yr recalc game from 1871-2024 a couple days ago, but I turned of minor leagues. Rico Carty ended up with a career 134 OPS+ in OOTP and his Real OPS+ on BBR is 132.

[Matt Arnold]

It will vary depending on what you have for your adjust/weaken settings.

I'm guessing that in 25, we used different cutoff thresholds, and so your 1962 ratings would effectively be the same as a 1964 single season replay. This year, we'll at least try to count some of those minor league seasons into the ratings total, hence his totals showing more like a minor league player.

[szathkey]

Makes sense…overall potential pretty comparable at least. I’m sure this issue is pretty similar in causation to the importing of historical players only rated at DH?

[uruguru]

Originally Posted by Matt Arnold

It will vary depending on what you have for your adjust/weaken settings.

I'm guessing that in 25, we used different cutoff thresholds, and so your 1962 ratings would effectively be the same as a 1964 single season replay. This year, we'll at least try to count some of those minor league seasons into the ratings total, hence his totals showing more like a minor league player.

I appreciate your response. Carty has 1924 professional PA (more than plenty) over the 5-year period, so adjust the weaken-hitters option has no effect.



In both 25 & 26, Carty's single-season projected slash line is 338/402/582, which is still quite a bit higher than 25's 5-year projected line of 296/360/502,


If you will indulge my research a bit... I looked at the single-season projected slash line for Carty in 60-64 and here's how they differed from v25 to v26. You can see the difference is very slight:

OOTP 25
YEAR PA AVG-OBP-SLG
------------------------
1960 67 140-177-220
1961 406 147-196-224
1962 397 165-202-253
1963 549 191-242-309
1964 505 338-402-582


OOTP 26
YEAR PA AVG-OBP-SLG
------------------------
1960 67 153-199-235
1961 406 149-208-222
1962 397 164-211-240
1963 549 200-259-315
1964 505 338-402-582

If you take those single season slash lines and weight them by PA (I have no idea what OOTP actually does), the weighted averages look like this, compared to what the game shows in the player editor:

OOTP 25
========================
WEIGHT AVG 213-264-350
------------------------
OOTP25 PROJ 296-360-502

OOTP 26
========================
WEIGHT AVG 216-274-347
------------------------
OOTP26 PROJ 196-253-307

So you can see that OOTP 26 is much much closer to the weighted average (which is good, assuming that MLB equivalencies are weighted equally to MiLB equivalencies). But the 26 projection is lower than the weighted average, which suggests that the MLB plate appearances are weighted lower than the MiLB appearances.


But there is a stark difference between the two game versions.