World...series!!!

[BigRapidsJackass]

Heltonfan said (re: the idea of matching up one of our lesser pitchers against Beckett in Game 1):

it's an incoherent idea. Anything you gain by matching Francis up with a lesser pitcher, you lose in the other game where you have to start someone else. Aside from the gains to be had by giving your best pitchers the most starts (I've said it before; Fogg should be the Game 4 starter, with Cook going in Games 3 and 7), it's a zero-sum game.

Which sounds logical. So I played around with some made-up numbers a little bit. Answer: it's not necessarily a zero-sum game.

Here's why. We'll make it easy. Assume each club is using 3 starters, not 4.

Assume each starter always pitches a complete game. Assume that the starters are of varying quality, and they always produce one of 3 possible outcomes, and that each outcome is exactly as likely as any other outcome. Assume that "tie games" go to extra innings in the bullpen, and that each club wins exactly half of these extra inning games. Assume no such thing as errors. All runs are earned. In fact, all runs allowed are solo Home Runs allowed. (This isn't really important to the scenario, but I don't want people getting stuck on ERA.) So we have:

Beckett: 1, 2, or 3 runs allowed per 9 innings. In other words, a 2.00 ERA guy.
Schilling: 3, 4, or 5 runs -- a 4.00 ERA guy
Dice-K: 3, 5, or 7 runs -- a 5.00 ERA guy

Francis: 2, 3, or 4 runs -- a 3.00 ERA guy
Ubaldo: 4, 5, or 6 runs -- a 5.00 ERA guy
Fogg: 4, 5, or 6 runs -- a 5.00 ERA guy

Match them up the normal way: Ace vs. Ace, #2 vs. #2, #3 vs. #3. There's 27 discrete results. Looking at it from the Rox perspective:

5 wins, 6 ties, 16 losses. Using our 50% tiebreaker formula, the Rox go 8-19.

Now let's match them up differently:

Beckett vs. Fogg
Schilling vs. Ubaldo (that one stays the same)
Dice-K vs. Francis

The 27 "games" result in the following:

8 wins, 3 ties, 16 losses. After the ties are broken, the Rox record is: 9.5 wins, 17.5 losses.

So under my thought experiment, "optimizing" matchups just got the Rox 1.5 more wins over 27 games.

Now, you may say that my numbers don't fit a real-world scenario. Well: (1) that's not the point. I am trying to show that different matchups could, under the right scenario, make a difference; (2) these numbers aren't wildly out of line. Everyone's ERA is between 2.00 and 5.00. Boston has an ace, a #3 type, and a #5 type starter here. We have a #2 type and two #5 types. Not the real world, but not inconceivable.

So the point is this: it's not "incoherent." It could make sense. Given the real world, correctly playing the matchup game probably would increase odds of taking a 3-game series only by a trivial amount. It would probably be overwhelmed by other factors.

But it's not necessarily a zero-sum game. Therefore, it does bear some thought. If managers labor over batting order (another thing that makes only a trivial difference), why not labor over starting pitcher matchups?

[Roxpert]

Jackass, that's some good serious thinking you did there. I intuitively felt the same way.

Tonight, Spilborghs is our DH and batting 9th. Now, Clint Hurdle should never be questioned for his Midas-like decisions, but this doesn't make sense on two counts:

1. Spilly is a better defender than Holliday, so why not use Matt as the DH and have Spilly man the Greee Monster? Is Clint trying to massage egos?

2. Spilly is a better hitter than Torrealba, so why not bat Spilly 8th and Yorvit 9th? Again, is this due to keeping egos intact?

I know that neither of thse nitpicks are likely to make the difference between a win and a loss tonight, and there are myriad other factors that will determine tonight's game outcome, but c'mon. IF this were the regular season, Spilly would be taking the field and batting no lower than 8th, while Matty would be getting a day off from leftfield to rest his legs. I suppose Hurdle wants to show off Holliday's all-around game on this huge stage. If he truly thinks Yorvit batting 8th and ahead Spilborghs gives us a better chance with Beckett, I don't quite know what to say. Clint must be playing "the hot hand" in the postseason with that decision.

[hiaspire]

Originally Posted by Roxpert

I suppose Hurdle wants to show off Holliday's all-around game on this huge stage. If he truly thinks Yorvit batting 8th and ahead Spilborghs gives us a better chance with Beckett, I don't quite know what to say. Clint must be playing "the hot hand" in the postseason with that decision.

I don't think it has anything to do with "showing off" Holliday. They are here to win, not anything else. Many hitters don't like DHing and feel like they are out of the game almost. Holliday isn't used to being a full-time DH like Ortiz. He's one of the leaders of the team and I'm sure wants to be out there pulling his weight and not just sitting in the dugout most of the game.

I think there certainly is a good case to be made to optimize the defense and DH Holliday (or Atkins) on paper, but then you are also taking guys you are counting on most out of their normal rhythm. They've already had a long layoff, and I think it is wise to get everyone back in there and playing in their normal roles as we get started.

Torrealba doubters have been eating it all post-season, and will continue to do so. I don't see any reason to demote him as he's been huge. Spilly also is much more of a top of the order type player and fits naturally in with Taveras and Kaz as they can play some small ball at the bottom of the lineup as it turns over in hopes of generating opportunities for the big bats. Lots of AL teams use a leadoff-type guy in the 9th spot knowing that once the batting order gets going he's going to have a very similar role to the top table-setters.

[BigRapidsJackass]

I'd say the curious thing is not DH'ing Hawpe and letting Spilly play RF.

Hawpe is, by any measure, our poorest outfielder. Spilly has played quite a bit in RF, and probably would fit best there. Actually, I think Holliday-Taveras-Spilborghs is one heck of a defensive outfield.

As for batting order: at the risk of infuriating Aspire by adopting his own theory, I'd let Taveras bat 9th as a kind of "2nd leadoff man." KazMat did well in the leadoff spot, and (at the risk of infuriating Heltonfan) I'm quite sure the stats show that Tulo was much more proficient in the 2-hole this year. Yes, I'm suggesting that Holliday and Helton afford him "protection." You can smite me with your SABR swords now.

EDIT: I checked the stats.

Tulo batting 2nd: 297/366/517, 883 OPS, 14 HRs in about 323 PAs
Tulo batting 7th: 294/359/445, 804 OPS, 8 HRs in 262 PAs

Meaningful? Probably just coincidence, but I'd take a shot and let him hit 2nd. What's there to lose? After all, even if it isn't meaningful, at least you'd give Tulo a shot at getting a 5th plate appearance every night. Since he's been a much more productive hitter than the alternatives (Kaz Mat or Willy T.), I'd rather give him the extra plate appearance.

I don't really care in which order Spilly and Yorvit hit. I don't think it makes any difference at all.

[Nate505]

Dang....not a great first inning so far....

[indianadrew]

holliday should have read that ball better

[indianadrew]

are we gonna have to listen to dane cooks crap on every commercial break. please god no more frankTV

[indianadrew]

if there is a stolen base we all get free tacos at taco bell

[John Cocktoston]

Man, Beckett is dealing tonight. He's just been a beast this postseason...

I don't mind Spilly at 9 simply because it gives you Spilly, Taveras and Kaz in succession. I don't know how much you gain by bumping Spilly to 8 and then breaking them up with Torrealba?

Wouldn't you rather have Spilly, Taveras and Kaz leading into Holliday, Helton, Atkins, et al?

[Roxpert]

Ok, I'm sold on Spilly hitting 9th. AS IF that would make a difference in this game.

Morales wasn't the same pitcher coming out of the bullpen. Now, we don't have a long man we can trust in this series. It may have been wiser to keep Buchholz on the roster, and deactivate Morales instead. Of course, it's easier to see this with hindsight.

[hiaspire]

What a disappointing world series debut. I'm still holding out hope that once we get into the bullpen that our bats can at least start to roll and make a charge. At least something to build upon. Funny things can happen sometimes when a team thinks the game is over. It probably is, but we still have some time to get something going.

This team has been through bad times before, though. They've had devistating losses where it looked like they might be out of it, and they bounced back well. They've gone on road trips with multiple walk-off losses from the grasp of sure victory.

The Rox have pointed to some of those moments as a time when the team collectively got angry and knew they were better than that, and turned things around. I'm still holding out hope til we are all out of outs, but if we go down perhaps it fuels future improved efforts.

[Roxpert]

Now, I think Francona just isn't playing this right. With a 13-1 lead, he has the luxury of taking Beckett out after 5 innings and 69 pitches. That way, if he so chooses, he could come back with Beckett on 3-days rest in Game 4. Instead, Beckett pitched the 6th inning and is up to 78 pitches. I guess there is no thought on Francona's part to even give himself the option of using Beckett three times in this series, if it goes 7 games.

[Heltonfan]

After all, even if it isn't meaningful, at least you'd give Tulo a shot at getting a 5th plate appearance every night. Since he's been a much more productive hitter than the alternatives (Kaz Mat or Willy T.), I'd rather give him the extra plate appearance.

Right. This is all that matters.

Jackass, you've gotten me thinking about the optimizing matchups thing. I have to concede that you're right, at least in theory; it's not a zero-sum game. But your assumed ERAs for the pitchers in question are so ludicrous that it undermines the whole thought experiment. I suspect that if you did the study using a more sophisticated setup (where each pitcher has more than three possible outcomes, and where the projections being used are halfway reasonable), the benefit of matching up differently would almost completely disappear.

Edit: it occurs to me that one simple way to do the matchups is to use Pythag. Using your ERA assumptions, our chances of winning, with each matchup, are as follows:

Francis vs. Beckett: 30.8%
Jimenez vs. Schilling: 39%
Fogg vs. Dice-K: 50%

Total of 1.20 expected wins in three games, or 2.4 in six games (since they'll be matching up twice).

Flipping things around:
Fogg vs. Beckett: 13.8%
Jimenez vs. Schilling: 39%
Francis vs. Dice-K: 73.5%

Total of 1.26 expected wins in three games, or 2.52 in six games. So the advantage is .12 wins over the course of the series, which is about a third of what you came up with using your simplified model.

If we use each pitcher's 2007 stats as their projection, rather than using the absurdly simplified numbers, the advantage of playing matchups is cut down to .03 wins over the course of the series. And if we use projections, there's not any advantage at all, because Beckett and Dice-K project equally well.

Morales wasn't the same pitcher coming out of the bullpen. Now, we don't have a long man we can trust in this series.

"Now"? As if to imply that, had he happened to pitch a scoreless inning, he'd all of a sudden be trustworthy? This isn't a hindsight issue; the guy's just not an MLB-caliber pitcher yet.

[Roxpert]

Originally Posted by Heltonfan

"Now"? As if to imply that, had he happened to pitch a scoreless inning, he'd all of a sudden be trustworthy? This isn't a hindsight issue; the guy's just not an MLB-caliber pitcher yet.

Well, not exactly. Tonight didn't tell me anything I didn't already know, specifically that Morales is a fragile and fidgety young STARTING pitcher, and when you place him in a role to which he isn't accustomed for the WORLD SERIES, he isn't apt to be trustworthy. Buchholz, on the other hand, has been a stabilizing force in the bullpen, and NOW we don't have him. That's the context in which I used the word "Now" in my sentence. You don't really think I am so dumb as to make that deterimination on Morales' outing tonight, do you?

(And I believe I wrote earlier today that I would have kept Buccholz on the roster. Now, we don't have him.)

[Dante Bichette]

Tough one tonight. I expected us to struggle with Beckett, but I also expected Francis to be much better.

Like others have mentioned, I am very surprised Francona left Beckett in as long as he did. He could have removed him after 5, with a 12 run lead, and easily brought him back for game 4, and then regardless of what happens in game 4 he would be available for a possible game 7 since there would be no tomorrow at that point. Facing Lester instead in game 4 should be to our advantage.