Correlation btwn individual players and nfbc finish

beausole · Post by **beausole** » Fri Feb 05, 2010 1:37 am

It's been awhile since I've posted or monitored the board much, I have to say that the level of discussion and depth of information and analysis this year far exceeds what I remember it being (without of course losing any of the 8th grade humor so needed from the many of us decades removed from it).

One thing I haven't seen on these boards, and I apologize if it's been done and I've missed, is anyone taking all the NFBC finishes in the draft format for a year (main event, satellite, slow draft, and anything else I've missed), or even just the main event teams, and doing a simple averaging of the place of finish for each team owning a given player.

I know a thing or two about stats and know that, even with many dozens of leagues of info for a given year, this would be subject to significant confounding and ample size issues. (The primary confounding variable being the 29 other players on each team.) My thought is that this would come out somewhat in the wash if you threw enough teams data at this. I think there would be a lot of obvious outliers and errors, and, relatedly, you'd be looking at very minor differences (tenth of a decimal point), but I'd find it interesting to see where teams that drafted a guy like Reyes last year (burned high first round draft pick) finished v. all the teams that drafted Mauer (bonus first round pick at third/fourth round cost, or less, depending on where drafts were versus injury report) v. Reynolds (deeper round bonus) v. Pujols (lived up to high first round pick).

Wouldn't necessarily inform that much for this year (maybe some risk aversion type things?), but would be interesting to some. (And would inform whether people could fairly justify a bad season on a single pick like Reyes.)

I'd do it but I don't have all the data. (Time would also be a slight concern.)

CC's Desperados · Post by **CC's Desperados** » Fri Feb 05, 2010 2:22 am

I personally don't see a lot of value in this information because last year is last year. I do know of one owner that probably has this information. I'm also pretty sure he's looked at it for the top twenty teams or so overall.

beausole · Post by **beausole** » Fri Feb 05, 2010 2:53 am

No, I agree it's not worth that much going forward for the individual players. It may have some value for player types. I'd also find it interesting. The guy that may have has the common players for the top 20 teams? That wouldn't be a bad list, as would the common players for the bottom 20 teams (Jeff Suppan).

CC's Desperados · Post by **CC's Desperados** » Fri Feb 05, 2010 2:57 am

I'll make a call and see if he will give up some information.

CC's Desperados · Post by **CC's Desperados** » Fri Feb 05, 2010 10:48 am

I made the call. He said he has suspended that part of his research. He deemed it time consuming and worthless information in the end. Sorry I couldn't help.

beausole · Post by **beausole** » Fri Feb 05, 2010 10:59 am

Thx much CC. Most definitely time consuming, and unlikely to be hugely useful. If however you ran it and saw that every speed guy on which a high draft pick was spent corresponded to a lower than average finish, regardless of whether they performed, or same with saves, or something else, or that teams that took Hanley rather than Pujols finished consistently (80 percent of the time) better, I'd find that useful. (I wouldn't necessarily share it, but would find it useful.) I'll make a note to myself to take the top and bottom 50 rosters in the overall at the end of this year and spend a few hours seeing if it shows any of these nuggets. If it does, I'll share.

CC's Desperados · Post by **CC's Desperados** » Fri Feb 05, 2010 3:14 pm

Originally posted by beausole:

Thx much CC. Most definitely time consuming, and unlikely to be hugely useful. If however you ran it and saw that every speed guy on which a high draft pick was spent corresponded to a lower than average finish, regardless of whether they performed, or same with saves, or something else, or that teams that took Hanley rather than Pujols finished consistently (80 percent of the time) better, I'd find that useful. (I wouldn't necessarily share it, but would find it useful.) I'll make a note to myself to take the top and bottom 50 rosters in the overall at the end of this year and spend a few hours seeing if it shows any of these nuggets. If it does, I'll share. Good Luck

Lawrence Liebman · Post by **Lawrence Liebman** » Sat Feb 06, 2010 1:37 pm

I tried to look at something like this and tried to focus on a "bushel of players" as opposed to isolating individual players. I looked at 35 hitters and 17 pitchers that offered "superior value" at their draft slot. I could have expanded the list further, but wanted to focus on guys that would have been drafted. I also limited the analysis to slow drafts (fixed roster, so the results wouldn't be influenced by sometime doing better in free agency) and teams finishing in the top 2 in the standings.

My conclusion was that it is important to hit upside picks, but because the whole portfolio matters, there was not a definitive correlation.

Here was the distribution (small sample, you'll see):

The first number shows the number of players from my list of overachievers that were on the winning team. The second number is the count of teams with that many players:

3 - 3

4 - 1

5 - 6

6 - 5

7 - 3

8 - 5

9 - 1

CBSSports.com also did an analysis by position on what players were most likely to be on winning teams. But, I believe the leagues they used in their analysis were all 10-team leagues. It is posted on their site.

I believe you can recover from a bad early round pick, but I don't believe any of the 24 teams in my sample had Reyes, though they did have other "bums".