Info on Blog

Friday, January 30, 2009

Travis Ishikawa and the Curious Case of the Platoon Label

Whenever I see any assessment of Ishikawa as a prospect, I usually see him listed as a future platoon candidate. I've accepted this assessment because he does strike out a lot and have had his troubles just hitting, let alone hitting lefties. However, now that he's at the major league level, the risks of making a bad assessment rises because before he could have just been one of a million failed prospects, but now he could be our starting 1B.

So I looked at his career splits over the past 4 seasons in the minors and they aren't that lopsided:

vs. LHP: .256/.341/.435/.776, .323 BABIP, 27.7% K%, 10.9% BB%
vs. RHP: .269/.356/.512/.868, .325 BABIP, 28.3% K%, 11.1% BB%

In fact, looking at his peripherals, his BABIP, K%, BB%, and BB/K are all basically identical, suggesting that the main difference between him against LHP and him against RHP is that he hits for much less power. And vs. LHP, he has more GB%, less LD%, so perhaps someone can work with him on that, plus give him a lot of BP against that LHP we have on staff for that ever since Barry got him.

The worse thing is his drop in power:

His XBH% is 47% vs. RHP, 38% vs. LHP.
His ISO is 243 vs. RHP, 179 vs. LHP
His HR/AB is 18.8 vs. RHP, 29.7 AB/HR vs. LHP

Still, he is slotted to hit 7th and the average 7th place hitter hit .259/.326/.406/.732, which is not that high a threshold for him to be MLB-average for us.

The funny thing is that everyone is saying that Ishikawa should be platooned, but Pablo Sandoval actually has a worse platoon split:

vs. LHP: .267/.320/.371/.691
vs. RHP: .324/.360/.490/.850

A Possible Estimation

Acknowledging that results vs. LHP tends to jump around a lot, one way to see how he might do in 2009 is to take his MLE vs. RHP in 2008 plus his MLB 2008 vs. RHP, and subtract from it the difference between his career MLE RHP and LHP to get a rough idea of what he might bat in 2009:

vs. RHP-08: .279/.326/.514/.840, 0.305 BABIP, 21.2% K%, 67% BB%, 20 HR in 401 AB
vs. LHP-MLE: .267/.310/.415/.725, 0.302 BABIP

And according to minors career, vs. LHP he has very similar peripherals.

With there being roughly 3 AB vs. RHP for every 1 AB vs. LHP, his overall batting line would skew more towards his RHP line, which is not that shabby for us, that would be roughly 800 OPS. I would love to get 800 OPS at 1B for us, we haven't seen that for years now.

Not sure how to factor HR, but with that many against RHP, even if he only hit 3 in 130-150 AB against LHP, that's a 23 HR season. When's the last time we could say that about a 1B on the Giants?

Comparison with Other LHH 1B in Majors

Not that I think he is anywhere near on par with any of the below hitters, but perhaps he can aspire to getting close to there, as there are some pretty ragged RHP/LHP splits at 1B in 2008, but they are just THAT good against RHP:

Ryan Howard:
vs. RHP: .268/.366/.601/.966
vs. LHP: .224/.294/.451/.746

Justin Morneau:
vs. RHP: .310/.401/.527/.928
vs. LHP: 284/.324/.453/.778

Carlos Pena:
vs. RHP: .280/.418/.576/.994
vs. LHP: .190/.302/.352/.654

Adam LaRoche:
vs. RHP: .282/.357/.513/.870
vs. LHP: .241/.298/.467/.765

Adrian Gonzalez:
vs. RHP: .320/.405/.585/.991
vs. LHP: .213/.287/.387/.675

Prince Fielder:
vs. RHP: .295/.401/.554/.955
vs. LHP: .239/.313/.420/.733

Career:

Ryan Howard:
vs. RHP: .304/.412/.652/1.065
vs. LHP: .231/.314/.471/.786

Justin Morneau:
vs. RHP: .294/.370/.529/.899
vs. LHP: 257/.301/.438/.739

Carlos Pena:
vs. RHP: .262/.374/.520/.894
vs. LHP: .226/.310/.442/.752

Adam LaRoche:
vs. RHP: .277/.345/.503/.848
vs. LHP: .254/.319/.448/.767

Adrian Gonzalez:
vs. RHP: .297/.366/.523/.888
vs. LHP: .248/.308/.427/.734

Prince Fielder:
vs. RHP: .292/.387/.565/.953
vs. LHP: .250/.330/.463/.793

Again, not that he will consistently do the above during his career, but as Ron Shandler says, once you show that skill, you own it forever, and he showed a pretty good skill against RHP in 2008. However, the key is to be able to duplicate that over and over again, meaning you somehow unlock how to bring out your skills regularly. Given how close his LHP hitting has been in recent years in the minors, on an overall basis, I would just give him the 1B job to see what he can do, whether he can put it all together finally.

2009 Season (and Beyond?)

As much as I would like to see a winning season, I still think it is more important to figure out what we got with the prospects we got. The 2009 MLB season should be about figuring out what Fred Lewis, Pablo Sandoval, Jonathan Sanchez, Sergio Romo, Alex Hinshaw, Travis Ishikawa, Nate Schierholtz, and Kevin Frandsen can do, heck throw Tim Lincecum and Matt Cain on the list too, they have something to prove too.

While it would be interesting to see if Phelps has any value, so that perhaps we can trade him, he's already over 30 years old, anything he shows will most probably start getting worse one year soon, whereas Ishikawa is still only 25 years old for 2009, which is still young for a player, and which Giants fans seem to forget, given how long he's been around as a top prospect. He still hasn't even reached his prime peak years yet.

Given the above analysis, I would just give Ishikawa the whole season (as long as he isn't hitting worse than 1B did for us in 2008: .248/.313/.371/.684) to see if he can progress again in 2009 as he did in 2008. If he can hit near to his MLE for 2008 in 2009, that would be very valuable for us, as 1B has scuffled for years now on the Giants and was not even on par for an average SS, let alone 1B.

If he is able to do that, that would set us up at 1B for a few more years and not have to feel like we need to rush Villalona up to the majors and just give him a year at a level at a time, instead of rushing him up like the D-backs did with Justin Upton. Still, if Villalona is able to rise one level a season, he would reach the majors in 2012 at the ripe old age of 21.

8 comments:

  1. Nice one OGC.

    Ishi will be very interesting this year. Say he fumbles early in the season, will they send him down and mix-up the infield around (assuming Pablo @3rd and Ish @1st) to Pablo to first and play Uribe at 3rd?

    Call me crazy, but I think The Jesus could bump Ish out, assuming Guzman's D has improved. from left to right: Frandsen-Renteria-Guzman-Pablo.

    Maybe I dont like Ishi from seeing him in the past... I dont know.. I just dont think very high of him.

    In my book, for him to play, Ishi would have to produce in every possible way for me not to play him.

    ReplyDelete
  2. Ishikawa have no more options, they can't send him down without exposing him to waivers, and I would assume that another team will pick him up.

    I like Ishikawa from seeing him in the past. He was in SJ and he CRUSHED a pitch over the mile-high netting they had in right field to protect the cars there from errant homers. It was AWESOME!

    I think because of his defense at 1B, I think he'll get more rope before he gets pulled. However, I think any extended period under 700 OPS would get him released, bring in the next candidate.

    I don't see the big deal with Guzman. He hit 14 HR in 342 AB in 2008 in AA. Ishikawa hit 16 in 171 AB in 2008 in AAA.

    Guzman hit .351/.405/.546/.951 cumulative, but with a .392 BABIP when his BABIP is .332 for the past four seasons. His stats are expected to regress to his mean, putting him back closer to his .293/.361/.469/.830 over the past four seasons.

    Ishikawa hit .299/.377/.578/.955 cumulative, with .323 BABIP when his BABIP is .325 over the past four seasons. This represents advancement in his skills whereas Guzman lucked out in 2008.

    And Guzman's defense sucked no matter where or what he played, except for LF. He was particularly bad at 2B and 3B previously, particularly 2B recently, costing his team about 1.5 wins with his poor defense.

    Ishikawa has been slighly plus most seasons, and he was particularly plus in 2008, his defense was worth 1.5 wins for his play in Connecticut prorated for a whole season, probably even better in Fresno.

    So not only did Ishikawa outhit Guzman in 2008, but he was worth at least 2 wins above Guzman defensively, perhaps up to 3 wins.

    Looking at Ishikawa's MLB defense in 2008, prorating it out using Fangraph's stats, his defense was worth 0.5 wins over a whole season (or roughly so).

    And most projection systems on Fangraphs peg him for .750-.800 OPS, and that would be great to have at 1B after so many years with much lower.

    ReplyDelete
  3. You're right, I did not factor him being out of options.

    In your post, you mention that it would be interesting to see if Phelps has any value. So would you like to see Phelps/Ish platoon v. LHP/RHP? Or are you indicating that you want Ish to play everyday to see how he does against LHP? If so, then there really is no way to get any value for Phelps while on the bench. Does Phelps still have options?

    ReplyDelete
  4. His split at Fresno is more dramatic than what you shown here. I don't have access to his CT data for 2008, but I remember seeing that the LHP/RHP split delta was quite big also. To be fair, if I were to cherry pick data, I would subtract his April (he had a horrible April of 2008). I think he may have out performed Dallas Macpherson from May to September, despite playing in much more pitcher friendly parks. Again I don't have access to data. But I think he'll be in the top 5 among minor league players in hitting statistics from May to September. It puzzles me that some posters in the various blogs were willing to trade Sanchez for Macpherson, but are not willing to give Ishikawa a chance.

    ReplyDelete
  5. Why anyone with a computer has access to the data: it's right here http://minorleaguesplits.com/cgi-bin/pl.cgi

    Ishi's a pretty hard case to figure out. Generally the more data the better a picture you can draw, but Travis' last four years have been so schizo in general and even more so in terms of splits, that it becomes really difficult to figure out what they mean. In his '05 year, which was a fine season in the Cal League, he actually performed better against RHP than against LHP. Over the next two years of scuffling in general, he was just about as terrible both ways. But then during his breakout year of '08 he displayed an extraordinary platoon split (including a truly horrid .522 OPS left-handed in CT). Again, one season's splits are way too small a sample size to tell you much, but the dramatic changes from one season to the next over the last four years, really muddle the picture much more than a cumulative stat line of those four years shows.

    ReplyDelete
  6. Thanks for the link. I guess I am still too lazy to dig out the April data. I recall that he had a horrific April, and much of the at bats against LH was in April. After that, it seems to me that they stop playing him against LH pitchers. Especially at Fresno, Harper got the playing time against LH. Therefore, his poor split may be skew by the fact that most of his at bats vs LH were during the period when he was bad in general. I suspect that had he been given the chance to play regularly vs LH pitchers in Fresno, his splits would improve. By the way, this is a lousy way to develop players. Instead of hiding his weakness by platooning in the minor league, they should given him as much at bats as possible vs LH so he can work on improving the weakness.

    ReplyDelete
  7. This is a good post. I was extraordinarily frustrated by Bochy's absolute refusal to let Ishikawa start/hit against LHP during his call up in September last year. The Giants wasted all kinds of ABs on Richie, McClain, and Ochoa last year when the Giants were totally out of contention. The Giants desperately needed to evaluate five players above all: Ishikawa, Schierholtz, Sandoval, Burriss and Velez. They did okay with getting Sandoval ABs, but I wish more of them would have been behind the plate so they could have evaluated him at C. Burriss got plenty of action, although more of it should have been at short because Ochoa looked offensively pathetic from day one and should never have gotten all those ABs. Schierholtz only got enough ABs once Lewis had surgery, and then he got hurt, so we don't half as much as we should about him. However, the failure to give Ishikawa enough ABs was inexcusable because Richie and McClain sure didn't need to be evaluated. He needed to start against RHP and LHP. In fact, especially against LHP to see what would happen. The Giants utterly failed in what should have been their only objective with Ishi in late August and September, to evaluate whether he can make hack it in the MLB, not to mention whether platooning will be necessary. I think Bochy gets the blame for this with his veteran worshipping ways.

    ReplyDelete
  8. About Phelps, what I was trying to convey was that while it would be interesting to see if we can set up some trade value in Phelps, I think it would be more valuable to the Giants to give Ishikawa the full time job and see if he can do it. If he can, he would be very valuable.

    About the varying splits, what I've observed with LHP and RHP splits is that while the platoon advantage side is relatively steady, the other side varies widely year to year, most probably because of the small sample problem. That's why I suggested looking at Ishikawa's 4 year overall MLE splits, to get an overall sense, once randomness is factored out a bit, what the suggested difference is for him. Obviously, my supposition above is predicated on my theory being correct. :^)

    The Giants wanted to end 2008 with a better record than 2007, when they had Bonds. They barely accomplished that. It's the balance between present and future needs that causes the dichotomy between what any of us wants for the Giants relative to what they do.

    There's nothing wrong with vets if you are trying to win for this season. There's some psychic benefit to finishing better with your young players than you did with Bonds, one of the greatest offensive weapons ever, even in 2007.

    ReplyDelete