OOTP Development- Player Churn

injury log · 12-10-2008, 07:19 PM

I was noticing, looking back over my past drafts in a solo fictional league, what seemed like a surprising number of draftees who had made it to the Majors. The league is set up to be as close to real life MLB as possible - same minor league structure, hidden players are on so the draft is not the only source of big league talent, 25 round draft (which seems to be about the number of draftees a real life MLB team will sign each year), AI evaluation settings at the defaults. Because I knew the players, it was pretty quick to go through 22 years of drafts to see what proportion of players pan out. First, the numbers who actually make the Majors, by draft round (it's a small sample from the top of the draft, because my team ended up being pretty good):

1st round: top 12 picks: 6/6______ 100%
1st round: picks 12-30: 20/25_____80%
2nd round: 16/22 _______________73%
3rd-5th rounds: 29/67___________43%
6th-10th round: 33/110__________30%
11th-25th round: 92/330_________28%

Additionally, I classified each player based on the quality of his career, and tallied up the number of 'useful' players from each draft - guys who at the least had long careers as bench players, or at least two years as above average regulars. The numbers:

1st round: top 12 picks: 6/6______ 100%
1st round: picks 12-30: 14/25_____56%
2nd round: 8/22 ________________36%
3rd-5th rounds: 12/67___________18%
6th-10th round: 11/110__________10%
11th-25th round: 28/330_________8%

On average, each draft yields 8.9 guys who see the Majors, and 3.6 guys who make a substantial contribution.

Of course, the number of guys that develop and make the Majors in my own team's system will be influenced by my own promotion/demotion decisions, and a proper study would look at an AI team's drafting- it was a lot faster to do by looking at my own team, which is the only reason I've done it this way. I did look over a couple of AI drafts, though, and the results were pretty much exactly in line with the above.

Now, looking over the BA 2003 Draft Database, and admittedly the jury is still out on a few of the HS draftees, it seems that most teams sign about 25 players, between 3 and 6 sniff the Majors from each team's draft class, and a smaller number than that actually make (or seem remotely likely to make) an important contribution. I might compile exact data at some point. In any case, OOTP's numbers seem markedly high.

This gets to one of the aspects of player development that I don't think has received enough attention. The game has gotten close to getting Tango curves right, but there hasn't been much data collected about how well OOTP models career arcs: is there enough variation year-to-year; do players reach the Majors at the right age; is there enough variety of career arcs; do the right number of players reach the Majors from each draft class; are career lengths right. The data above suggests that far too many draftees are making it to the Majors, which suggests two possible issues:

too many draftees are developing the skill set to actually perform at the Major League level;
too many big league vacancies are occurring, giving these players the opportunity to amass playing time. My guess is that, because of the severity of OOTP talent hits, OOTP is churning through big league players: they develop, get promoted to the bigs, play for a year or two, take a talent hit because of injury or randomness so severe as to turn them back into minor leaguers, then get replaced by a recent draftee.

Anyway, it's not an altogether systematic study, but I wanted to at least get a starting point and put it out for discussion. It would be interesting to compare with real life data on the odd chance that someone has collected it for a real life MLB draft.

Isura · 12-10-2008, 07:23 PM

Talent variation is a factor. Like you said with 100 talent randomness there's a lot of 1-2 year wonders. Good research, I look forward to reading more.

RonCo · 12-10-2008, 08:13 PM

Unless something changed in the last patch or three, the average OOTP player's career is considerably too short (so, yes, players are churned through).

Afino · 12-10-2008, 08:18 PM

I wish there was two seperate talent randomness figures.

One for established major/minor leaguers. This could be set VERY low (10? 25?) to emulate Tango curves better.

One for prospects. This could be set very HIGH (125? 150?) to not make top 5-10 picks automatic major leaguers (as you've illustrated and I've also seen) and a couple more lower round picks become superstars.

12-10-2008, 07:19 PM	#1
injury log Hall Of Famer Join Date: Apr 2007 Location: Toronto Posts: 9,162	OOTP Development- Player Churn I was noticing, looking back over my past drafts in a solo fictional league, what seemed like a surprising number of draftees who had made it to the Majors. The league is set up to be as close to real life MLB as possible - same minor league structure, hidden players are on so the draft is not the only source of big league talent, 25 round draft (which seems to be about the number of draftees a real life MLB team will sign each year), AI evaluation settings at the defaults. Because I knew the players, it was pretty quick to go through 22 years of drafts to see what proportion of players pan out. First, the numbers who actually make the Majors, by draft round (it's a small sample from the top of the draft, because my team ended up being pretty good): 1st round: top 12 picks: 6/6______ 100% 1st round: picks 12-30: 20/25_____80% 2nd round: 16/22 _______________73% 3rd-5th rounds: 29/67___________43% 6th-10th round: 33/110__________30% 11th-25th round: 92/330_________28% Additionally, I classified each player based on the quality of his career, and tallied up the number of 'useful' players from each draft - guys who at the least had long careers as bench players, or at least two years as above average regulars. The numbers: 1st round: top 12 picks: 6/6______ 100% 1st round: picks 12-30: 14/25_____56% 2nd round: 8/22 ________________36% 3rd-5th rounds: 12/67___________18% 6th-10th round: 11/110__________10% 11th-25th round: 28/330_________8% On average, each draft yields 8.9 guys who see the Majors, and 3.6 guys who make a substantial contribution. Of course, the number of guys that develop and make the Majors in my own team's system will be influenced by my own promotion/demotion decisions, and a proper study would look at an AI team's drafting- it was a lot faster to do by looking at my own team, which is the only reason I've done it this way. I did look over a couple of AI drafts, though, and the results were pretty much exactly in line with the above. Now, looking over the BA 2003 Draft Database, and admittedly the jury is still out on a few of the HS draftees, it seems that most teams sign about 25 players, between 3 and 6 sniff the Majors from each team's draft class, and a smaller number than that actually make (or seem remotely likely to make) an important contribution. I might compile exact data at some point. In any case, OOTP's numbers seem markedly high. This gets to one of the aspects of player development that I don't think has received enough attention. The game has gotten close to getting Tango curves right, but there hasn't been much data collected about how well OOTP models career arcs: is there enough variation year-to-year; do players reach the Majors at the right age; is there enough variety of career arcs; do the right number of players reach the Majors from each draft class; are career lengths right. The data above suggests that far too many draftees are making it to the Majors, which suggests two possible issues: too many draftees are developing the skill set to actually perform at the Major League level; too many big league vacancies are occurring, giving these players the opportunity to amass playing time. My guess is that, because of the severity of OOTP talent hits, OOTP is churning through big league players: they develop, get promoted to the bigs, play for a year or two, take a talent hit because of injury or randomness so severe as to turn them back into minor leaguers, then get replaced by a recent draftee. Anyway, it's not an altogether systematic study, but I wanted to at least get a starting point and put it out for discussion. It would be interesting to compare with real life data on the odd chance that someone has collected it for a real life MLB draft.

12-10-2008, 08:18 PM	#4
Afino Hall Of Famer Join Date: Sep 2008 Location: Buffalo, NY Posts: 3,828	I wish there was two seperate talent randomness figures. One for established major/minor leaguers. This could be set VERY low (10? 25?) to emulate Tango curves better. One for prospects. This could be set very HIGH (125? 150?) to not make top 5-10 picks automatic major leaguers (as you've illustrated and I've also seen) and a couple more lower round picks become superstars. __________________ GUBA: Moscow Enforcers

12-10-2008, 07:23 PM	#2
Isura Major Leagues Join Date: Oct 2008 Posts: 361	Talent variation is a factor. Like you said with 100 talent randomness there's a lot of 1-2 year wonders. Good research, I look forward to reading more.

12-10-2008, 08:13 PM	#3
RonCo Hall Of Famer Join Date: Aug 2003 Posts: 10,502	Unless something changed in the last patch or three, the average OOTP player's career is considerably too short (so, yes, players are churned through).