Yesterday we found that retaining many of one's youngsters does not ensure success on the field. Being a Phillies fan, I know that sometimes running your homegrown talent out year after year is only a good idea if you homegrown guys are good—remember Steve Jeltz?

But what if we could filter out those homegrown players that burn a hole in your team's roster. I reran the numbers looking at just those players who had at least 100 Win Shares in their careers. Here are the results sorted by the percentage of time the average 100-WS prospect spent with his first club:

So how well does that correlate to winning percentage? Not at all actually (coefficient of -0.0842). Maybe 100 Win Shares just isn't enough. I ran the data again using a 200 Win Share cutoff:

I like what I'm seeing toward the top of table. Does it correlate any better? Ever so slightly, but not really (0.1527).

Ok, as a last ditch effort I will limit the data just to players with 300 or more career Win Shares, basically a Hall of Fame-caliber player. So how'd we do this time?

We improved significantly (0.4327 coefficient), but I'm still not happy.

Maybe looking at the total number of players produced who eventually collected the desired Win Share total is enough. Maybe it doesn't matter how much time those actually spend with the team. Let's see. I ran the numbers and actually it correlates pretty well for 100 Win Shares (0.7662). But what's odd is that it goes down significantly as the Win Share cutoff goes up, eventually approaching what we saw above at 300 WS (0.6538 for 200 WS and 0.4332 for 300).<.p>

Let's try looking at just the average number of years the player spent with the team that developed him. Those numbers don't correlate as either of the other two (0.3418 for 100 WS, 0.2470 for 200, and 0.3469 for 300), and it takes an odd dip in the middle.

So what have we learned? It's most important to develop credible major-leaguers in large numbers and it's best to retain the best players for the longest amount of time. That seems to make sense, sort of the Branch Rickey player development approach when he revamped the Cards' minor league system. Quantity and quality.