[go: up one dir, main page]
More Web Proxy on the site http://driver.im/
THE BOOK cover
The Unwritten Book
is Finally Written!

Read Excerpts & Reviews
E-Book available
as Amazon Kindle or
at iTunes for $9.99.

Hardcopy available at Amazon
SABR101 required reading if you enter this site. Check out the Sabermetric Wiki. And interesting baseball books.
Shop Amazon & Support This Blog
RECENT FORUM TOPICS
Jul 12 15:22 Marcels
Apr 16 14:31 Pitch Count Estimators
Mar 12 16:30 Appendix to THE BOOK - THE GORY DETAILS
Jan 29 09:41 NFL Overtime Idea
Jan 22 14:48 Weighting Years for NFL Player Projections
Jan 21 09:18 positional runs in pythagenpat
Oct 20 15:57 DRS: FG vs. BB-Ref

Advanced

Tangotiger Blog

A blog about baseball, hockey, life, and whatever else there is.

Minors_College

Minors_College

Monday, November 12, 2018

Prospect Valuation

?Terrific stuff from Craig at Fangraphs.  Loved the whole idea and execution.  This is what sabermetrics is about.

***

One of the interesting points was the present value of future WAR (not to be confused with the present value of future dollars).  Craig applies an 8% annual discount rate which seems high to me.  The question to ask yourself is this: if you had a 4 WAR player today, what would you need in future WAR (say 3 years from today) in order to give him up? Clearly, it has to be more than 4.  Is it 4.1?  4.5?  5?  6?  Prima facie, I'd say 6 is far too high.  5 seems too high as well and obviously 4 is too low.  The answer should be close to 4.5?  If you think of it in terms of future WAR 1 year down the road, maybe it would be 4.1 or 4.2?  And two years down the road it would be 4.3 maybe 4.4?  

So, what kind of discount rate would that imply?  A 4% discount rate would suggest that you would trade a 4 WAR player for this kind of WAR player, if you wait these number of years:

  • 1: 4.17
  • 2: 4.34
  • 3: 4.52
  • 4: 4.71
  • 5: 4.91

If you tried different numbers, you will  find that 2% seems low and 7% seems high.  So, what is the marketplace answer?  I don't know.  It would be easier in other sports because they trade draft picks.  You can try to look for it in MLB by looking for veterans-for-prospects trades, but there's going to be other considerations there that may prevent us from getting at a good number.

Anyway, great stuff all-around.

Sunday, April 09, 2017

How can we handle survivorship bias?

?@mitchellichtman

I don't know if this is an epiphany.  One of the issues with looking at the forecasts of poorly forecasted players is that when one of them has a terrific start, he will keep playing, while those who have poor starts won't have a chance to keep pitching.  And then we end up weighting the players based on playing time, giving more weight to the guys who got luckier.

So, when I saw this post from MGL, talking about Guthrie, I had a thought: what if MGL gives us the list of pitchers who he says is AA-level (or worse).  And then we look at the FIRST game they play in MLB.  And only the first game.  Indeed, I would even say: let's look at the first 9 batters they face in MLB.  Or even, just the very first batter.

I presume there's some 100 pitchers every year that MGL thinks is AA-level or worse who pitch each year, facing at least one batter.  He does this for the last 10 years of his forecasts, and now we have 1000 PA of what should be AA-level or worse.  Did they match MGL's forecasts?  Or did the teams know something he didn't?

(58) Comments • 2017/04/18 • Forecasting Minors_College

Wednesday, June 29, 2016

Regression amounts for college stats

?Terrific stuff from Jeff.  

Interlude: Just to give you some context, in MLB, Marcel adds 240 PA of league average stats, meaning that at 240 PA, half of what you see is real, and half is random variation.  (If you did it by component, it would require much less regression for K/PA and much more for BABIP.)

Jeff is showing you need 1000 PA at Division 1 play to regress half toward league average.  For JUCO, it's only 400 PA.  

Saturday, January 30, 2016

How much is a top prospect worth?

According to Chris Mitchell, the ?top 100 prospects are expected to generate over 600 wins during their pre-free agency years, or an average of 6 WAR.  A pre-free agent player's career probably costs about 30 cents on the dollar, meaning that 2 of those 6 WAR are being paid at the full price, and the other 4 WAR is surplus.  (A 6 WAR player is the 36th best prospect according to Chris.)

Suppose you have a veteran player, one year to go until free agency.  Those guys get paid 60 to 80 cents on the dollar.  If your superstar player is expected to get 6 WAR in his last year, 4 of those wins are being paid the full price, while the other 2 WAR is surplus.

You can see therefore that the average top 100 prospect is worth about twice as much as a superstar free agent in his last year.  According to what we can infer from Chris anyway.

(1) Comments • 2016/01/30 • Minors_College

Monday, September 14, 2015

Can tackle football in high school survive?

?I am pretty much dumbfounded that this high-profile a sport tied with concussions can exist in high school.  Isn't it just one lawsuit away from being eradicated?  I presume there used to be high school boxing and I also presume it no longer exists? 

In Canada, we don't have this issue because the leagues are NOT tied in to the high school.  Which would seem to be the path that's going to happen in USA.  My question: why hasn't it happened already, and when will it happen?

Now, as for college: because we are dealing with adults, it would have a lesser burden to consider.  But still, once the high school domino falls, the college football program would be in jeopardy(*) as well, wouldn't it?

(*) I always spell it as jeapordy initially, and the spell-checker saves me. I think my spelling is more consistent with the pronunciation.

Monday, February 09, 2015

Imagine if players and teams had to court each other?

?This is a fascinating process in terms of making sure everyone is happily married, and how it applies to doctors and hospitals.

Imagine therefore that this would replace the amateur draft.  In addition, a player comes with his own price tag, of which he can neither increase nor decrease.  And teams set up their max budgets as well.  Then let the (automated) matching begin.

Wednesday, February 04, 2015

Removing the draft

?Dave proposes removing the draft altogether, and simply giving each team caps on new players.  This is something we've talked about in the past.  Whereas we normally tie it to win% (the lower the win%, the more you can spend on new players), Dave suggests tie-ing it to team payrolls.

Three years ago, this was my proposal.

(8) Comments • 2015/02/05 • Minors_College

Tuesday, December 30, 2014

Minor league components for forecasting

?Great job by Chris.  I like the little thing he did by using WAR through age 28.  I do something similar.  I also like that he says:

Even in just deciding whether a [low minors] player will crack the majors, most of the projections fall somewhere between between one percent and 15 percent.

For the "statistical significance", setting aside the term itself, there should be no gaps.  Basically, each level is not independent, and it doesn't make sense that the BB rate is "significant" for A- and A+ but not A, that K% is significant at every level except A+.  It's just a sampling issue.

A very well laid-out article.

(1) Comments • 2015/01/18 • Minors_College

Tuesday, December 16, 2014

Trade value of prospects

?Kevin and Steve do a good job of coming up with valuations by prospect ranking.

Friday, December 12, 2014

Should minor league ballplayers be exempt from Fair Labor Standards Act?

One side suggests... YES!

For whatever reason, Baseball America did not provide the counterpoint.  I don't know anything about the FLSA, but let's say that "apprentices" or "trainees" are exempt.  Now you have to define what that means.  I can't imagine a 5-year or 10-year "trainee" could exist. 

In addition, I can't imagine the "trainee" to be the primary service being provided to the paying customer.  You go to a minor league game, you pay your 20$ for the ticket to watch... 18 trainees compete?  Every game, for a season?

In addition, the "trainee" is bound to an organization?

HOWEVER... signing bonuses.  Those signing bonuses is what should save MLB.  Basically, the idea is "I'll give you a one million dollar signing bonus, and we'll give you a monthly stipend for up to six years."  Now, THAT I can buy.  Now, it's very clearly an investment with an apprentice.  That kind of relationship seems more than plausible.  The minor leaguers need to acknowledge this signing bonus.

Now, for any minor leaguer who did NOT receive (more than a nominal) signing bonus: then I agree, the FLSA should apply to them.  All you gotta do is make sure the signing bonus is at least 10 x 10 x 180 x 6 = about 100,000$.  That's 10$ an hour, 10 hours a day for 180 days for six years.

All that other stuff in the article, about the sky is falling, all that is b.s.  I think the main argument is in the above paragraph.

Rise and Fall of the Phillies Draft Empire

A simply terrific article and research by our buddy (and Phillies fan) Matt Swartz.

Perhaps the most astonishingly bad time period was 2004 through 2007, since most of the players drafted during this time period across the league already have started augmenting their big league clubs. The Phillies’ draftees from those years have accumulated just 9.6 career WAR while the league average across all teams has been 76.1 WAR. This discrepancy explains why the Phillies have been so far out of first place during the last three years. It is simply not possible to have such colossal draft failures and still compete in baseball today.

 

Friday, December 05, 2014

Geographic Bias in Draft

?Dan and Alex do a good job in looking for geographic bias in the draft.

On a related note: Didn't Matt Swartz have a similar article a few months ago on this topic?

(2) Comments • 2014/12/08 • Minors_College

Wednesday, November 26, 2014

Blueprint for revising MLEs? Or is an MLE just an intermediary step?

A good explanation from Dave and Kevin as ?to the issues and considerations with MLEs.  They get to all the touchpoints, and give a good overview.

And the first comment on the article by Mike Green was also on point: if MLEs are just an intermediary step to forecasting, do we need to think about MLEs to begin with?

That is, does it payoff to create an MLE for Jeremy Reed and his 2003 AA half-season?  Or, do we simply want the true talent equivalency of that (whether equivalent for AA opposing talent or MLB opposing talent)?

Think of it this way: let's say that Jeremy Reed went from one AA league to another AA league.  Would we simply create an "exchange rate" of 1:1, and so, maintain his .409/.474/.591 slash line?  That means his good luck carries over.  If so, then if he went from AA to AAA, do we continue to carry his good luck forward?

Suppose we do.  Then what?  Well, then we'd need to regress the observed-but-exchanged data.  That's what Mike is getting at.  If no one is really interested in the intermediary step, then do we need to take that step?

Well, we might, insofar that we don't have a good handle on how to handle such biased data, and Dave and Kevin properly point out.

So, it's an interesting process to go through, and far too little is discussed of MLEs.  And far too much is discussed on less useful aspects of baseball analysis.

Sunday, August 31, 2014

Both HIGH SCHOOL pitchers go the distance in a 50-inning 700 pitch outings

?Wow,

Saturday, July 26, 2014

Timeouts (finally) coming to baseball!

?The Atlantic League, but that's a start.  I've been clamoring for nine timeouts per team for ANY reason (stepping out of the box, changing pitchers, mound conference, possible injury, etc).  Any extra is a "technical" and requires an automatic called ball or called strike on the batter as the case may be.  The Atlantic League is limiting it to 3 DEFENSIVE timeouts, with the penalty of called ball if they exceed it.  This is great!  Now, if they can enforce 6 OFFENSIVE timeouts, then I'd be very happy.

The courtesy runner for catchers is interesting.  Courtesy runners were more common in the "golden age".  I like the idea at least for experimentation.

IBB: automatically called. I like it, so at least we won't be subjected to its absurdity, of making everyone go through the motions.

Umpires asked to enforce a rule already in the book... as if only some rules are enforced!  That's the biggest silliness.  "We have a rule!"  "Nah, I won't enforce that one."  "Well, now we are specifically telling you that's not one of the rules you are allowed to ignore!"  I'd rather just go with defined timeouts.

Thursday, July 24, 2014

MLBPA, Astros, NCAA

Murray Chass gives us all the deets, including:

Another is the Astros’ release of medical information about Aiken’s elbow. Since he was not their player, they had no authority to disclose confidential medical information in violation of the privacy rule of the federal HIPAA law.

(1) Comments • 2014/07/26 • Minors_College

Tuesday, June 10, 2014

Luck in the draft

There's two types of luck.  One in which no one participating has any knowledge of the subject matter, and so, making selections is "pure luck".  A non-sports fan choosing the winner between Kings and Rangers would be pure luck. 

Another is in which EVERYONE participating has the same maximum knowledge, and so with no differentiating skill among the participants, any selection is again "luck".  If say Larry Bird and Reggie Miller are the best free throw shooters from the 3-point line, and after 1000 attempts, they both hit 70%, then choosing between one or the other is going to be luck.

Pizza looks into the draft and basically sees no evidence that a team has skill in picking players RELATIVE TO THE OTHER 29 TEAMS.  It's luck, but only after every team has refined their process to their maximum.  If one team decides to abandon scouts altogether, then we WOULD see evidence of skill.

Anyway, I liked his approach.  What's interesting is that seeing results where the extreme teams have 50% more (or half as much) picks pan out relative to the average team is still not enough to say whether there is skill involved (beyond the skill that all teams have).  Basically, we'd have to observe a team having overwhelming success, in order for us to believe they have even a small level of skill.

?

Thursday, June 05, 2014

Inside the draft room

Tony gives us an inside look as to how it all works.?  And Jeff picks out passages out of an MLB memo as to what can and can't be done.

When Tulo was drafted

?Interesting story on Tulo.

Monday, June 02, 2014

Dirk Pants-on-Fire Hayhurst

Dirk has a story of how he lied himself into getting drafted.  This story is familiar though.  Did Dirk write about this already?

Page 1 of 4 pages  1 2 3 >  Last ›

Latest...

COMMENTS

Nov 23 14:15
Layered wOBAcon

Nov 22 22:15
Cy Young Predictor 2024

Oct 28 17:25
Layered Hit Probability breakdown

Oct 15 13:42
Binomial fun: Best-of-3-all-home is equivalent to traditional Best-of-X where X is

Oct 14 14:31
NaiveWAR and VictoryShares

Oct 02 21:23
Component Run Values: TTO and BIP

Oct 02 11:06
FRV v DRS

Sep 28 22:34
Runs Above Average

Sep 16 16:46
Skenes v Webb: Illustrating Replacement Level in WAR

Sep 16 16:43
Sacrifice Steal Attempt

Sep 09 14:47
Can Wheeler win the Cy Young in 2024?

Sep 08 13:39
Small choices, big implications, in WAR

Sep 07 09:00
Why does Baseball Reference love Erick Fedde?

Sep 03 19:42
Re-Leveraging Aaron Judge

Aug 24 14:10
Science of baseball in 1957

THREADS

November 12, 2018
Prospect Valuation

April 09, 2017
How can we handle survivorship bias?

June 29, 2016
Regression amounts for college stats

January 30, 2016
How much is a top prospect worth?

September 14, 2015
Can tackle football in high school survive?

February 09, 2015
Imagine if players and teams had to court each other?

February 04, 2015
Removing the draft

December 30, 2014
Minor league components for forecasting

December 16, 2014
Trade value of prospects

December 12, 2014
Should minor league ballplayers be exempt from Fair Labor Standards Act?

December 12, 2014
Rise and Fall of the Phillies Draft Empire

December 05, 2014
Geographic Bias in Draft

November 26, 2014
Blueprint for revising MLEs? Or is an MLE just an intermediary step?

August 31, 2014
Both HIGH SCHOOL pitchers go the distance in a 50-inning 700 pitch outings

July 26, 2014
Timeouts (finally) coming to baseball!