Fangraph's KATOH top 100 prospect list

[ZBTMP]

http://www.fangraphs.com/blogs/katohs-top-100-prospect-list-for-2016/

 

Brewers well represented, including (2) recent additions higher than any scouting or prospect guru list. Might better explain our new more analytical minded front office additions....

 

I know you probably know this, but I’d like to reiterate that you shouldn’t think of this as “Chris Mitchell’s Top 100 List,” and certainly not “FanGraphs’ Top 100 List.” This is simply the output from a flawed statistical model that fails to take into account many of the factors that go into evaluating a prospect. As always, you should never choose between stats and scouting — or beer and tacos — if you don’t have to.

[MrTPlush]

The list is painful mess and inaccurate, but it is interesting both Flores and Betancourt are analytically liked.

[CheezWizHed]

This is simply the output from a flawed statistical model

 

That kind of says it all, doesn't it? [sarcasm]I guess if DS keeps acquiring people high on this list (and only high on this list), we should be concerned about his statistical capabilities.[/sarcasm]

[And That]

I sincerely doubt that the Brewers use KATOH whatsoever in their statistical analysis. That's not to say they don't use some models that may be similar - players who are young for their league that perform well if not spectacularly - but I think it's mostly a coincidence that Betancourt and Flores are where they are on that list. Now if they trade for a few more relatively obscure guys on that list, you might be on to something.

 

Stearns has said time and again that he values analytics and scouting more or less evenly. Until there is significant evidence of a weight toward one or the other, I'll keep my mind open to the process, even if the results are KATOH-esque.

[CheezWizHed]

I should've put that in blue, because I was only half serious.

[SRB]

Why do they even publish these lists?

 

Wasn't Brandon Diaz in the top 50 or something a couple season ago?

[ZBTMP]

Last year was the first year it was published and Diaz was not in the top 200;

http://www.fangraphs.com/blogs/katohs-top-200-prospect-list/

 

I'll defend the system somewhat. Obviously, there is a lot of volatility in projecting based off of minor league stats no matter how good your methodology is. That being said, the methodology still has correlation and it isn't completely junk.

 

Everyone (including the projection system) knows Orlando Arcia and Corey Seager are good prospects and they are valued as such. But valuing prospects without eye popping stats, but do things at an age that correlate to big league success seems essential to gaining any kind of advantage from analytics. The fact that Betancourt and Flores graded out in the 98th percentile of all minor leaguers in such an analysis is noteworthy.

[And That]

Diaz was in BP's top 100 two years ago, I think.

[sheetskout]

I think some of you are taking this list too seriously. The author is up front about this being an experimental model. This is more about a statistical model/algorithm in development than actually predicting prospect outcomes.