THE BOOK--Playing The Percentages In Baseball

I looked up his context-adjusted lwts in my files.  Sometimes there is a large gap between OPS+ and lwts. I did not realize that he was such a bad hitter.  He really is.

His UZR’s are not all that special.  Looks like maybe an above average SS and a very good 2B, but nowhere near someone like Everett.

We should probably call him the luckiest player in baseball, to make even a few hundred thou.  He should be toiling in the minors somewhere making about 25 grand per.

Can you post his UZRs, along with his positional-teammates?  With a case like this guy (lots of data with and without), it would be interesting to see.

I understand that while at most positions a replacement level hitter is an average fielder, a replacement level shortstop is a -5 fielder.  I think that’s Nate Silver’s research.

A terrible hitter like McDonald doesn’t have to be otherworldly with the glove to hold a job, just be a good fielder.

He doesn’t have to even match replacement level in SLWTS (though I have him closer to that than Clayton) to keep a MLB job.

1. Some managers prefer a good fielder on the bench, even at the expense of a better all around player.

2. You can leverage a player like this.  Play him in late innings after your regular has hit for the last time.  Or in the games he starts pinch hit when he comes up in a crucial situation.

I quickly checked my db of active players and Peter Bergeron comes very close to meeting tangotiger’s criteria.

I think he’s the only guy who is still playing pro ball that is a real contender, however. Cesar Izturis also has a shot at joining this exclusive club.

That 2004 of Izturis made people believe he could hit.  Over 60% of his career starts are as leadoff or #2.  He may very well have the largest gap of fielding to hitting, while being a horrible hitter.

Having both Bergeron and Chavez in the same OF was tough on Montreal.

Neifi Perez is pretty much the same player as Cesar Izturis.

Bergeron was a guy who was supposed to hit.  He kept getting chances until he convinced management that it wasn’t going to happen.

Right - none of those guys zre really in the same class as McDonald, Salazaar, etc. because of their usage patterns.

I guess adding a “past the age of 24” caveat to the criteria would weed out most of the players who kept getting plate appearances because they had the potential to hit.

If only Mario Mendoza could field…

I was wondering how many guys are there like Mendoza, who didnt pass the fielding part.

Coincidentally, I went to high school with John McDonald (he was a year ahead of me).  As I recall, he wasn’t even a great hitter at that level, and there were about three or four players at our school who I thought were better players than he was.  I was absolutely stunned when I first saw him in a game five or six years ago (his name is common enough that it never occurred to me that it was the same person until I actually saw him on TV).  He was an awfully slick fielder, though, easily the best fielder I’ve ever been in a game with.

So, yeah, probably one of the luckier careers in history, considering he hasn’t had exceptionally good hitting numbers at any level of organized baseball.

Angel Salazar was an absolute wizard in the field. Yet another of the great shortstops out of Puerto Rico. Yeah, he couldn’t hit his way out of a paper bag. What, did he have 1 homer in his career or did he get to 2? Regardless, great player.

Signed, 2/3.8, implying a WAR of 0.7.

This is similar to Adam Everett, who signed a slightly better contract.  MacDonald is a slightly worse hitter than Everett (if that’s even possible), and slightly worse fielder.  Does that combination even allow him to be a MLB player?

For his career, he’s a -3 wins per 600 PA player.  Jays fans rave about his fielding, though UZR doesn’t share that rave.  He’ll be close to winning a Globe Glove this year.  That’s worth +2 wins above average (fielding and position).  So, he’s -1 win relative to average, or +1 relative to replacement.  He’s not a full-time player, so, giving him a 2/3.8 deal is a fair price.

By the way, MacDonald, from 2000-06, has 3.2 A/9 at SS (178 games) and 3.3 at 2B (102 games).  His teammates at SS (1106 games) were 2.97 and at 2B (1181 games) were 3.07.  In both cases, he made +.23 assists per 9IP more than his teammates (behind the similar pitchers, similar parks, yada yada yada).  In 162 games, +37 more plays, or +30 more runs.

Tango, the assist figure for his teammates looks awfully low.  Normally shortstops will have a few more assists than second basemen.  McDonald’s teammates included a mid-30s Vizquel, Peralta, Guillen, Russ Adams and Aaron Hill.  I wouldn’t be shocked to learn that this grouping would be below average, but not as much as your figure would suggest.  +30 runs would be better than Ozzie Smith, and I have a lot of doubt about that number.

Right, that’s 37 plays better than his teammates.  His teammates could very well have been -10, and he’d be +20.

On top of which, it is only 300 games, so there’s enough uncertainty there.

From 2000-2006, I’ve got MacDonald with 569 assists, while his teammates on the field had 4800 outs (includes strikeouts).  So, per 27 outs, that’s 3.20 assists.

His teammates: 3280 assists by SS out of 29852 outs, or 2.97.

+20 would be in the Dave Concepcion/Cal Ripken peak range.  That’s very, very good. 

In ‘07, McDonald has had 3.37 assists per 9IP, while the other shortstops on the club have had 3.35 assists per 9IP.  These things are very sensitive to the quality of the other shortstops on the club.  Had Clayton finished the year in Toronto, McDonald would have had similar results to 00-06, but with the arrival of Olmedo, the figures changed considerably.

Right, a one-year sample isn’t going to help us.  Even in the 2000-2006 sample, there were 18 different SS, with the following:
58% Vizquel
14% Adams
7% Perahlta
6% Guillen
6% Hill
4% Infante
6% rest

2.92 assists for Vizquel, and 3.04 for all the others.  Each a long way from the 3.20 that MacDonald recorded.

I had thought that the mid-30s Vizquel was league average or a little below, but that might be entirely wrong.  Vizquel did have a positive UZR for the time, and his Dial Rating (at least for 2006) was positive as well.  It’s strange that UZR and other measures of range (as well as the subjective view) should have such a different account of the relative fielding ability of Vizquel and McDonald than assists/9.