Jack Cust hit second Tuesday, a spot that manager Bob Geren said before spring training he was kicking around as a possibility. The acquisition of Cabrera makes that less likely, however. Cabrera has hit second much of his career and bats right-handed, making him a potential good fit there for Oakland

.

5. The value of taking second base/cost of the out, depends on what subsequent hitters do. Think of it this way: If there’s two outs, and the next hitter will only K, BB, or hit a HR, there is 0 value to being on second instead of first, and the cost of the out is high due to the possibility of a BB or HR. If there’s 2 outs and the next hitter definitely will not BB or get an xbh, the cost of making an out trying to steal is near 0, and the value of the extra base is very high. This logic applies to real less extreme cases to a lesser degree, but for those reasons. The 75% figure is an average of all situations that does not apply to individual decisions.

making it evident that Duke really should have started on three days’ rest to begin with.

Rajai makes it, no room for two backup OFers, Denorfia is stashed away again at AAA.

Whether or not he hits left-handed pitching as well or better than Sweeney is as relevant as how well Estelle Getty hits left-handed pitching, because neither are going to be an Oakland Athletic this year, unless somebody gets hurt.

The best contact hitter isn’t automatically the best overall hitter, nor is the best power hitter automatically the best overall hitter.

I value a hitter that can go to all fields and make good contact, and I’d probably have that guy second in my lineup after someone with a combination of speed and a better ability to get on base. After those two I want two or three hitters that are better at hitting for power, ideally one of the three being good enough at making more consistent contact in the 4-spot. I don’t want to have the middle of the order filled with power hitters who can’t do anything else very well.

After that, the lower part of the order may be more of a mix-and-match job, but I’d still like a good contact hitter 7th or 8th just to have a better chance of keeping a rally going that the top half of the order gets started.

None of this is scientific or researched but this is how my ideal lineup would probably go:

1 – one of the best on the team at getting on base, but the faster he is the better
2 – one of the better contact hitters and on-base guys
3 – power hitter, but also effective at contact
4 – power hitter, but also one of the best at hitting overall and getting on base
5 – power and/or contact hitter
6 – good overall hitter, best power after the rest
7 – someone left over from the middle of the order, may even be a little worse than #8
8 – hitter with decent speed, good ability to make contact and get on base
9 – whoever’s left

I’d probably expect some disagreement with this.

Correct, long relievers usually come in when your startin pitcher tanks it. Of course, both team’s starters could tank it, then the game becomes a battle of the bullpens. But that’s the exception.

In the playoffs, you have an off day every 2nd or 3rd day, so your best relievers have more time to rest. Plus, there can be many days between series. In the playoffs, I think it’s very possible to get 4-5 innings per game out of your top 3-4 relievers. That still means they’d only pitch of an inning per day on average during a series, with extra time off between series to rest.

I think a very strong bullpen may be the most economical way to put together a team that is a very good playoff team. Hopefully, the A’s will get to test that theory this year.

I remember a Ken Macha interview going into the ALCS against Detroit where he said something like: “The playoffs are no different than the regular season. You give your ball to your starter and you let him go.” B.S. The Angels’s wouldn’t have won the World Series in 2002 if they didn’t get 18 1/3 innings in 7 games out of K-Rod and Donnelly. In the regular season, you’d only average 6 innings out of these two over 7 games.

Thus, it’s possible your top relievers are 3 times as important in the playoffs as they are in they are in the regular season. This bodes well for the A’s if our starting rotation can overachieve enough to put us into the playoffs.

Beimel me Billy!

Yesterday’s pre-game interview with Geren shows me how complex his decision making was. Korach asked: “What does your lineup look like?” Geren answered kinda like: “It depends on the pitcher. You may have a lefty who isn’t very tough on lefties…” He listed 3 or 4 more things he considered. There’s probably more. Korach said “Wow, I expected a simple answer like: Sweeney leadoff, Cabrera 2nd, Giambi, Holliday…”. Geren said: “That sounds pretty good to me.”

Colorado to Oakland is a pretty big deal. The other thing is that I think most projection systems tend to project star players at lower levels than they’ve been in the previous years, as is the case for Holliday. Just because there’s some regression to the mean and stars are pretty far from the mean. Both the chone and zips projections would be a substantially lower ops+ from his previous three years.

…so calling Ellis the best second baseman in baseball clearly isn’t correct. I can go on if you’d like (Pedroia though still early). Ellis has had 2 subpar offensive years in the last 3 years (putting up .326 OBP, .404 SLG). I really like Ellis. But I know how he’s produced over the last 3 years.

I see a lot of “my stats are so much better than your stats” comments. Here is the list of top players (position adjusted) in the last 3 years. What is yours? If your stat is better than mine, then show me some proof. This is a VORP list so no defense is included, but people overrate defense IMHO. I still think all these guys would be in the top 10:
84.7 Albert Pujols
75.0 Hanley Ramirez
71.3 Alex Rodriguez
68.4 Chipper Jones
67.2 David Wright
65.6 Miguel Cabrera
65.4 Chase Utley
64.1 Matt Holliday

Cust vs. Ellis: Cust was ~30 runs better on offense (per full year) over the past two years. 10 runs = 1 win, so he’s 3 wins better than Ellis on offense. Ellis gets 1-2 wins back by virtue of being a 2B instead of a corner Of/DH. Let’s just say 1 win
Second Base: +2.5 runs
 …
Left Field: -7.5 runs

Isn’t it just 1 win (2.5+7.5 runs is 10 runs = 1 win)? And saying it’s 1-2 wins is “stretching” the truth. If you need to stretch your truth possibly by a factor of 2, then I feel like you’re reaching.

It sounds like Ellis is getting double points for his position and his defense…and it feels like defense is being overvalued here (Bill James eye-balled defense as only 17% of the game). I’ll look into this soon, since I like Ellis, and I’d like to think he’s as valuable as Cust. I will be the first to say that I do not have a firm grasp on defensive stats. For now, I think we need to agree to disagree.

mikea, you wrote:

vorp has a variety of flaws that are complex and tedious to detail

I understand VORP does not include defense and I fully accept that. But defense is only about 1/4 of an offensive player’s value (obviously defense more important for C, SS, 2B, CF).

Can you point me to these articles that say VORP isn’t all that?

reading this through, it sounds like to me that giving him 1-2 wins back for being a 2nd base person is not an effective measure.

How I read this comment is: Each position is more, or less important to the overall defense of the team. Being a good middle infielder will have a greater multiplier effect on that individuals defensive skill (good or bad) than a corner outfielder. A bad defender at 2nd would be more harmful than a bad defender in left. Giving every secondbaseman the same 1-2 win upward departure seems wrong.

I of couse could be misreading.

mikeA thanks for all your replies. I realize how much time it takes. That said…

It’s wrong to give away points just for playing a position. You should give away points based on how well a player is at that position. Chase Utley should get the maximum number of positional adjustment points. Guys who either can’t hit well for a position shouldn’t get the same positional value as Utley (I thought about including fielding but they overrate fielding already). It looks like they multiply your ABs by times played at a position. If a guy is a bad hitter at 2B, there could be a mediocre hitting OF like Kelly Johnson or Skip Schumaker ready to take their place. So why give away so many points to a guy if he plays 2B poorly if you have a guy on your roster who can replace him?

It also seems like Fangraphs is giving points to tough positions in their Replacement column. With the highest Replacement values are Jimmy Rollins, Jose Reyes, Sizemore, Ichiro, Juan Pierre, Dan Uggla who are all middle infielders/center field types. It seems like they are burying their love for up the up the middle defense in the Replacement column as well.

Which gets us to the contradiction: “this middle infielder is so great because he can play SS and he can play LF”. One of the reasons he doesn’t play LF is that he can’t hit well enough to play LF. Everybody thinks fond thoughts of the SS who hones his skill over the years working at his craft of defense. Defense is only 17% of the game. Why don’t we hold in even higher regard a guy like Edgar Martinez who was in the batting cages the day after the season ended hitting for hours honing his craft. Edgar incorrectly gets marginalized because he is a DH. Edgar’s career .418 OBP is 22nd best ever. His OPS of .933 is 34th best ever. And yet, I suspect Fangraphs (and people in general) would say Edgar isn’t that good because he DH’d.

Which brings us to Edgar’s buddy Raul Ibanez. Raul Ibanez signed a $30 million, 3-year deal with the Phils. Now GMs don’t always make good decisions, and that may or may not be a good price. I think it’s a little, but it was signed earlier in the off-season before prices dropped. Ibanez hit .293 BA .358 OBP .479 SLG (not Edgar-esqe, but not bad). Ibanez also played alot getting 707 PA, which he should get credit for. Ibanez has been a very good hitter in an extreme pitcher’s park. But Ibanez is a bad fielder and he plays LF. Fangraphs has Ibanez as being worth (gasp!) $2 million ($3.5 million on chart adjusted to average 2008 salaries * 89.775 / (87.5 + 69.0) ). Thus, Fangraphs is saying 2008 Raul Ibanez is worth less than the 2008 version of Bobby Crosby, Rajai Davis, Jack Hannahan, and Travis Buck with Carlos Gonzalez worth just a bit less than Ibanez.

Would Billy Beane give Ibanez $10M per year. No way. He’s got the slightly better Cust locked up thanks to baseball’s indentured servant rules. But Beane would certainly snatch up Ibanez at $2 million and even more if starting a team from scratch.

Doesn’t it seem like Fangraphs is overvaluing defense and position not just a little, but a lot (factor of 2 or more comes to mind)? Am I preaching to a choir of me here?

Cust was one of the worst corner OFs, at 10-25 runs below average.

Well, we came up with a number that said Cust cost us 5.6 runs. It’s odd they have a such a wide range of 10-25…or a range at all…for a guy with no range. :)

Then, Ellis, as the best 2B in baseball the past few years, was 15-25 runs above average (among 2Bs) with the glove (again, per full season),

Well, I’m running out of steam on this discussion partly since it seems like I’m the only one who cares. I have to think that if Fangraphs is off for Cust they have to be off fairly equally for Ellis. Cust only played 70 games worth of innings, so maybe he clocked in at the low end number of 10. If we take Ellis’s low end number of 15, that’s a difference of 25 runs (which assumes Cust played. Cust is 34.2 runs better offensively (determined via Cust and Ellis lineups). Cust still wins barely 34.2 Cust offense – 20.6 Ellis defense – 12.5 Cust half LF/DH positional adjustment (and I think the positional adjustment is wrong). I contend that Ellis gets too much credit for his position since in 2008 he was 35th in VORP at 2B.

Interestingly the positional adjustment for Cust works out about the same if you say he played 1/2 time in LF and 1/2 time at DH (-18.1 = -5.6 defense – 7.5/2 – 17.5/2) or you double his defensive badness and say he played full time in LF (-18.7 = -5.6*2 – 7.5). So their positional adjustments could be correct for LF and DH at least (assuming a decent hitter). Of course, I’m not saying the positional values are correct (I think they could be off by a factor of 2). But it’s comforting that they say Cust does about as much damage defensively whether he is DH or in LF.