Player Creation/Maintenance

Player Creation/Maintenance

(Note: the following is a detailed description of how players’ ratings are generated for this league. It is not required reading; it is here for reference.)

In this league we use some players who were career minor leaguers and some who played in both the major and minor leagues, as well as Negro League players who never played in either the majors nor the minors. As a result, different methods are used to create different players’ ratings.

(I should mention at this point that OOTP now comes with databases that include Negro League and PCL players. Unfortunately that was not yet the case when this league started, so I had to create a majority of the players for this league myself. I still do that, rather than use the new databases, because switching would have been problematic for a number of reasons, not the least of which is that I’ve studied OOTP’s Negro League and Minor League databases a bit, and I’m not sold on them. They’ve adjusted the players’ ratings to reflect the level of competition, which is the right thing to do, but I don’t agree with the adjustments they’ve made. My methods may not be any more accurate than theirs, but mine follow a logic I can understand.

  • Negro League players and players who never played in the majors are “created” using the “Create Fictional Player” function. The statistical set used for a player’s ratings is determined by where he was playing in real life during the year we’re currently playing. Generally, if he was playing in the majors that year, his ratings will be derived from his MLB career; if he was playing in PCL, his ratings will be derived from his career totals at the highest minor league level (P.C.L./A.A./I.L.); and if he was playing in the Negro Leagues that year, his ratings will be derived from his Negro League career. We call these real-life stat-based ratings his “base” ratings (as distinguished from what his ratings might be at any given moment, since OOTP’s Player Development can and will alter them). Since most players played in different levels during their career, their base ratings can change from year to year.

  • Players’ ratings are derived from their career totals, not from individual seasons. A player’s “current” and “potential” ratings are initially set to be identical. (Exception: if the player enters the league as a teenager I downgrade his “Current” ratings a bit so he can’t dominate the league when he’s 18. Unless he really did dominate the league when he was 18.)

  • Players who played in the majors are imported from the database that comes with OOTP, and I reset their batting or pitching ratings so that they are based on their MLB career statistics.

  • OOTP’s Player Development (in all its randomness-embracing glory) is enabled, but we rein it in when it gets too squirrelly. We want to see PCL stars like Buzz Arlett, Frank Shellenback, Jigger Statz, and Tony Freitas have long and successful careers, just like they did in the real the PCL; same holds true for great Negro League stars like Oscar Charleston, Biz Mackey, Jud Wilson, and Willie Foster, all of whom played well into their 40’s. To me it would be no fun at all to see OOTP hit one of these guys with a career-ending injury when he's still young, or to reduce him to journeyman status while he should be in his prime. We also don’t particularly want Joe Nobody who had 135 at-bats in his career to turn into a superstar. So we only allow Player Development to take the players’ ratings so far before we zap them back to their base ratings. For details, go here.

  • I make the assumption that the overall quality of competition in both the PCL and the Negro Leagues was slightly inferior to that of the major leagues. As a result, when I create the Negro League and PCL players, I use their real-life stats, but downgrade them a bit.

Anyway, down to the specifics:

  1. Missing data: We have no strikeout data for minor league pitchers, or walks or strikeouts for minor league batters before 1941. There are other less important stats (HBP, WP, etc.) missing as well. For minor league players who had 100 or more AB or 100 or more IP in the majors, I use their MLB numbers to supplement the missing data. I have a book (The Pacific Coast League: A Statistical History, 1903-1957—heretofore referred to as the Green Book) that contains some of the missing data for selected players as well. Basically, I use the data that I have available to me. Where no data exists for a statistic, I use the major league average for that statistic.

  2. The following “penalties” have been assessed for each PCL player:

    • Hits, doubles, triples, home runs and walks for batters decreased by 10%; strikeouts for batters increased by 10%.

    • Hits, walks, runs and earned runs for pitchers increased by 5%; Strikeouts decreased by 5%.

      • I know it doesn’t make any logical sense to penalize batters by a different percentage than pitchers. The 10% seems to “work” well for batters; if you look at players who went from the PCL to MLB in the 20’s and 30’s, it’s striking how many of them there are whose PCL batting stats are about 10% better than their MLB batting stats. No such correlation exists for pitchers, however. As a group, the PCL pitchers’ hits allowed and walks allowed rates are already so high that penalizing them even 5% seems severe. We’ll see how it goes; I may adjust the penalty lower or higher at some point.

  3. The following “penalties” have been assessed for each Negro League player:

    • Hits, doubles, triples, home runs and walks and hit by pitch for batters decreased by 7.5%.

    • Hits, walks, runs, earned runs, home runs, wild pitches and hit by pitches for pitchers increased by 7.5%; strikeouts decreased by 7.5%.

  4. I am using the stats from baseball-reference.com for minor leaguers, plus the stolen base and pitchers’ strikeout data from the Green Book (for the players it includes; it’s not comprehensive).

  5. I am using the stats from seamheads.com for the Negro Leaguers, rather than using baseball-reference.com. There are advantages to this, but drawbacks as well:

    • On the plus side, the stats from seamheads appear to be far more extensive and accurate. Seamheads includes some statistics that bb-r does not, such as earned runs. As far as I can tell, seamheads does not contain obvious errors in the walk data (bb-r clearly does).

    • The drawback to using seamheads is that it is incomplete. The Negro League database at seamheads is an ongoing project; it is far from finished. They have, however, been running at a pretty good clip; they do a league/season at a time, and since they started in 2011 they have completed most of the work through the early ‘30’s (they don’t always do seasons in chronological order). I’ll update the data I’m using whenever seamheads posts a new season. Keep in mind that when you draft a Negro League player, his stats/ratings are subject to change. As a general rule, the older the player, the more likely seamheads has done all or most of his career; hence, the data for the older players is less likely to change.

    • I’m using only the stats from the Negro Leagues. Seamheads also has stats from Latin American leagues, in which many African-American players played, but I’m not using those.

  6. OOTP has made it fairly easy to create batters by inputting statistics, but its exact process for rating pitchers is incomprehensible. I won’t go into detail; you really have to try it yourself to understand how cryptic it is. Over the years I’ve been able to figure out how they arrive at some of the ratings. I’ve developed charts for Stuff and Control ratings, and I use a combination of those ratings and ERA to determine Movement ratings (which are supposed to be based on home runs allowed, a stat we don’t even have for 95% of the players). For most of the other ratings (individual pitch ratings, GB%, velocity) I refer to the OOTP version of the pitcher, and use the same ratings they gave him when applicable, and alter them when not (remember “my” version of a Negro League or PCL pitcher may be radically different than OOTP’s).

  7. Pitchers’ stamina ratings: Prior to each draft I make the following adjustments:

    • Pitchers with 300+ IP that year in the real-life PCL: Stamina rating of 20

    • Pitchers with 250-299 innings that year in the real-life PCL: Stamina rating of 18

    • Pitchers with 200-249 innings that year in the real-life PCL: Stamina rating of 16

    • Pitchers with 150-199 innings that year in the real-life PCL: Stamina rating of 14

    • Pitchers with 100-149 innings that year in the real-life PCL: Stamina rating of 12

    • Pitchers who pitched in the real-life PCL that year with < 100 IP: Stamina rating of 8

    • Star Negro League pitchers get a Stamina rating of 20. Negro League pitchers brought in as “Buddies” get a Stamina rating of 12.

      • These ratings are intended to encourage GMs to use the pitchers as they were used in real life, without making it mandatory.

  8. Commissioner override: Sometimes a player meets the real-life minimum requirement (100 IP or 100 AB) for eligibility in the league, but has a limited statistical record that, in the Commissioner’s opinion, would produce unrealistically good ratings for the player if he was rated purely based on his statistics. The Commissioner may choose to “dumb-down” this player’s ratings. This is to discourage GMs from using marginal players as if they were stars.

    The Commissioner might also, in rare cases, choose to “enhance” a player’s ratings to encourage his use. Typically this would be a player who was a long-time regular in the PCL but whose real-life stats are less impressive (perhaps misleadingly, due to park effects or other factors).

    Players affected by this Commissioner override will be marked with a (CO) next to their names on the Excel sheets.

  9. Fielding and baserunning ratings: There are some fielding statistics available online for both Negro League and PCL players, but it’s limited. I’m not sure how OOTP arrives at all of its fielding and baserunning ratings for MLB players; I assume they’re all stat-based, but I don’t know how the stats are applied to create some of the ratings. In general, I rate all the Negro League and PCL players as “average” for baserunning and fielding except in the following cases:

    • Sufficient statistical data exists to suggest a different rating; e.g., the guy stole a lot of bases, or had a huge number of assists, etc.

    • Sufficient written accounts exist that suggest the player was outstanding or outstandingly awful in some category

    • Common sense suggests something, e.g. the guy was an ordinary or subpar offensive player but held down a starting job for a significant period (this would suggest he was probably an above-average fielder)

    • I do use real-life fielding percentages for Error ratings, making allowances for the fact that fielding percentages in the Negro Leagues are, for reasons I don’t understand, generally a lot worse than fielding percentages in the majors or minors. As a result I made different Error Ratings charts for Negro-League and non-Negro League players.

      • If you feel you have some information about a specific player that runs contrary to a rating I’ve given him, speak up! I’m no expert on these players; additional information is always welcome.

    I use the following chart as a basis for defensive ratings:

    AVERAGE DEFENSIVE RATINGS


    IF Rng IF E IF Arm IF DP OF Rng OF E OF Arm C Arm C Abil
    P (avg. rating: 11) 5 4 5 4
    C (avg. rating: 11) 11
    13
    1B (avg. rating: 9) 6 4 7 7
    2B (avg. rating: 11) 11 11 10 12
    3B (avg. rating: 10) 11 10 13 10
    SS (avg. rating: 10) 14 11 13 13
    LF (avg. rating: 10) 10 9 9
    CF (avg. rating: 10) 14 12 10
    RF (avg. rating: 10) 10 9 11

    The chart is subject to change. I will continue to refer to OOTP’s MLB player ratings—however they may evolve—to form the basis of my ratings for PCL and Negro League players.

  10. Injury/durability ratings: I base these ratings on how much career the player has ahead of him; if he’s at the beginning of a very long career I give him a very low rating; if he had a short career or is near the end of a longer one I give him a higher one. This doesn’t guarantee anything; some players are going to get injured more often than they should no matter what rating you give them, while others will be inexplicably impervious to injury.

  11. I monitor players to make sure Player Development doesn’t transform them into something that barely resembles what they were in real life. I don’t think it’s necessary or desirable to reset all of the players’ ratings each year; some will improve and some will decline; I like that their career arcs won’t be completely predictable. But when the change is severe I step in and reset them back to their original settings. For the gory details of what exactly has to happen for me to reset a player, go here.

  12. Career-ending injuries/early retirement: Recent versions of OOTP seem to have curbed the frequency of CEI’s (career-ending injuries) considerably, but just in case, I stand at the ready to un-do CEI’s in some cases. I allow the game to assert its bloodthirsty will on “ordinary” players, but I make players who had substantial (10+ years) careers immune, up until two years prior to the real-life end of their careers.

    • OOTP also decides when players retire. Some retire earlier than they did in real life, others later. I exercise a manual override here, too, and prevent the game from making players with substantial careers retire until two years prior to their real-life retirement.

  13. Missing years/late retirement: If we play long enough to get to the WWII years, players who missed years due to military service will miss the same years in our league. Aside from that, players who missed seasons in real-life will be eligible to play those seasons in our league, if only because we can’t know for sure that they really did miss those seasons (neither bb-r nor seamheads are infallible).

    • Since OOTP doesn’t always make players stop playing at the same time they did in real life, some will be able to extend their careers beyond their real-life retirement, and in some cases, beyond their real-life lives (OOTP loves to hurt players, but it never kills them). I am okay with this; it’s an alternative history, so the circumstances that caused players’ deaths in real-life have been altered. We may assume that our players have access to better medical care and are perhaps a little more careful about the company they keep. That being said, we don’t let them go on indefinitely; we forcibly retire any player once he has played two years past his final real-life season.

More essays

Home