THE BOOK--Playing The Percentages In Baseball

OK, here is an analysis of today’s lineups.

If you read the lineup analysis in the Game I thread, you will see that Tony has a pretty good lineup, even batting Jay, the sole lefty, second versus a LHP, Wilson, and Berkman, who is awful versus lefties, 5th.

Washington, however, appears to have an awful lineup - about as bad as you can get without being ridiculous.  At least according to my sim.  His lineup cost Texas 1.18% if win percentage for that game, which is 1,9116 wins for an entire season, which is enormous of course.  His biggest mistakes appear to be batting Andrus, probably the worst hitter in the lineup, second, and Napoli, the best or second best hitter (depending on whether the SP is RH or LH) in the lineup, 7th.

Here is an analysis of TEX’ lineup today.  We’ll see if it is as bad as yesterday’s.  Remember, this is according to my sim, which uses my up to date projections, platoon projections, base running, relievers, pinch hitters, etc.  Of course, I cannot comment on or know how each hitter likes each slot in the lineup.  I have always said that if a hitter is not comfortable in an optimal slot, you probably don’t want to move him there.  But who knows?

Here is tonight’s TEX lineup against Garcia, the lefty:

I KINSLER 2B
E ANDRUS SS
J HAMILTON LF
M YOUNG 1B
A BELTRE 3B
N CRUZ RF
M NAPOLI C
C GENTRY CF

With that, TEX scores 3.539 rpg and wins .5241.

If we change it to this:

I KINSLER 2B
N CRUZ RF
J HAMILTON LF
M NAPOLI C
M YOUNG 1B
A BELTRE 3B
E ANDRUS SS
C GENTRY CF

We get TEX scoring 3.564 rpg, .025 runs better than with their actual lineup, and they win .5305, or .65% more, which is 1.05 win per season, again, not a trivial amount.  Also, I’m not sure why TLR has Punto in there against a RH starter when Schumaker is now on the roster.  Punto really is an awful hitter and he drags down the entire lineup.  He appears to be a much better defender though.  You can certainly leverage their strengths by starting Schumaker and them putting in Punto if you get a lead.  Of course with Punto in the starting lineup, you have a much better hitter off the bench, at least versus a RH reliever.

I’ll substitute Schumaker for Punto and see what we get.

TEX scores 3.660 which is .096 more runs, due to Schumaker’s worse defense.  However, SLN scores 3.232 rpg, which is .138 rpg better than with Punto (3.094), and TEX wins .5234, which is a net gain of .7% for SLN, or 1.15 wins per season.  So, it appears that TLR does not have an optimal lineup out there.  Again, I would start with Schumaker and switch to Punto some time in the middle or late innings, if the Cards take a lead.  (My sim does not do defensive substitutions.)

You would think that McCarver, after catching for 20 years and then broadcasting for another 20, would be able to tell the difference between a slider and a change up, especially when they show the hand and the spin in super slow motion close up.  But, I guess not…

A whole win (for a year) from that is mostly making sure Elvis Andrus doesn’t bat second is pretty damn huge, and we’re not talking about how many runs you get for batting the pitcher behind or ahead of some fellow Punch-and-Judy hitter. Which one of you wrote the bit in The Book about how the whole project being worth it if managers stop batting bad hitters second?

I’d normally give a break on Punto since I’ve always liked Punto, but that is pretty damn big.

You would think that McCarver, after catching for 20 years and then broadcasting for another 20, would be able to tell the difference between a slider and a change up, especially when they show the hand and the spin in super slow motion close up.  But, I guess not…

No doubt.

If you show it to Mccarver with the infrared camera, he would be able to tell it was a slider/cutter.

I’d have definitely hit for Colby Lewis to lead off the sixth. Lewis has to project as a slightly below average pitcher three times through the order. And the Rangers have a few pitchers in the bullpen who project as well above average (Feliz, Adams, Ogando, Oliver). Of course, a pinch hitter is going to be about .20 to .25 runs better per PA than Lewis is.

Go to Gameday,after every pitch, place your finger on the screen, and then see the pitch plot 10 secs later. RU ever off by more than 6”?

http://www.mlb.com/mlb/gameday/index.jsp?gid=2011_10_20_texmlb_slnmlb_1&mode=gameday&partnerId=LR_gameday#gid=2011_10_20_texmlb_slnmlb_1&mode=gameday

Exception for first pitch to get your bearings, but after a while, you get used to it.

Classic. Gotta love the base hit just after Joe and Tim explain that Beltre’s not seeing the ball.

I have ESPN Radio and GameCast going, and there’s a delay on the radio. so I see the whole at bat before the announcers babble about it. That one was funny, indeed.

Yup, Phil, I would have PH for Lewis in the 6th. No manager would though. Yeah, Colby is a barely above average starter in the first place.  Third time through the order?  Any reliever is better.

Pinch hit for Punto in the 7th?  I don’t see any reason why not.  If he brings in Schumaker (or Moreland) to PH, will Wash put in a lefty? I don’t think so.

McCarver also called the first pitch to Cruz (I think) in the 3rd or 4th a “deep slider”. I didn’t check the Gameday, but it was definitely a fastball.  Again, they showed the ball coming out of the hand and the spin…

Remember what I said last night about Ogando’s pitch?  You have 2 outs and first and third in a tie game.  You throw hard.  Where do you want to tend to throw your fastball?  Up and in!  Not down and away or even up and away.  Too easy to dink one into RF, which is what he did both times.  You need to avoid the base hit and you don’t mind the walk or the HBP.  You pitch inside!  Not all the time of course, but there is where you want to tend to pitch. Maybe one of the reasons Scioscia did not like Napoli was things like this.

If you are trying to avoid an extra base hit (or a hit batter) but don’t mind giving up a base hit, then you do the opposite.  You avoid the inside part of the plate!

Tango, #6, I never thought it was that difficult to plot a pitch by eye, despite what some of the pitch f/x guys say.  Especially if you watch a lot of games and you mentally adjust for the slightly offset camera angle (some broadcasts are more centered than others)…

Are those opinions without evidence?

Yes, I am trying to be funny.

Almost everything Freese and Craig have hit has been middle away. I thought for certain both guys would get pounded inside with hard stuff, especially low and in sliders. I was wrong.

Charles, I would love to hear Washington’s reasoning for Napoli batting so low.  It’s not like he has been cold or has a low batting average.  His defense of Andrus batting second is going to be the usual - he wants a fast guy that can bunt, etc. in the 2 hole, like most managers. He doesn’t understand the downside to giving him so many PA, the fact that he has no power and hits behind a guy who gets on base a lot (Kinsler), and the fact that if a guy is a good base stealer, you don’t want to have power guys following him…

Circle, good one.  Those are the things for which we all value your opinion - like where they should be pitching those guys, given the situation. I think the situation is at least as important as the batter’s strengths and weaknesses.  For example, in the late innings when you have a small lead, and you don’t want to give up a home run, you see most pitchers stay away, away, away, against power hitters, even if they would normally pound a batter inside.

Not much going on in today’s game (not much to screw up) given that both pitchers were getting outs most of the time and there were few base runners.  I’m pretty sure, though, that in the next inning, we have a 50/50 chance to see a screw up…

MGL: right, this doesn’t have to be a theoretical discussion.  It’s practical.  Just try it!

Funny story… just now, I had my finger on the screen, waiting for Gameday to catch up.  The third pitch already came in, and I was wondering when the 2nd pitch would plot.

Well, while I was watching TV, the pitch had already plotted… right under my finger!  I moved my finger out, and there it was.

Anyway, I’m challenging everyone out there to JUST DO IT.  Try plotting it.  Once you get your bearings, and you use the various clues, like the catcher, the batter, it becomes a cinch.  (Though, there haven’t been any huge breaking pitches when I was doing it.)

In a vacuum, if you were to stop the video at the moment the ball hits the glove, and zoom in, so that you can’t see the batter, then yeah, sure, that’d make it much much tougher.

But, in the reality of what I see?  It’s easy.

And, there’s simply no way I can believe that BIS would need to adjust even 1 pitch’s ball location, based on what I’ve been plotting.  Like I said, I nail them all within 6 inches, probably averaging 2 or 3 inches at most.  No way should BIS override that.

Now, global calibrations, sure, they can do that.  But not pitch by pitch.  PITCHf/x should be considered defacto perfect there.  (Calibration issues aside.)

MGL,

I know you’ve often talked about players running out grounders or apparent home runs. I’m curious what your thoughts are on Pujols watching his fly-out. And also whether your opinion is different tonight than it would be on a random Thursday night in June.

Pinch hit for Punto in the 7th?  I don’t see any reason why not.  If he brings in Schumaker (or Moreland) to PH, will Wash put in a lefty? I don’t think so.

Moreland’s a Ranger.

Charles, I would love to hear Washington’s reasoning for Napoli batting so low.  It’s not like he has been cold or has a low batting average.  His defense of Andrus batting second is going to be the usual - he wants a fast guy that can bunt, etc. in the 2 hole, like most managers. He doesn’t understand the downside to giving him so many PA, the fact that he has no power and hits behind a guy who gets on base a lot (Kinsler), and the fact that if a guy is a good base stealer, you don’t want to have power guys following him…

Listening to Wash’s pressers on a fairly regular basis, I don’t think he’d actually say this. At this point of the year, he’d definitely say that it’s worked for them and he’s not going to change it now. That’s his spot, it’s worked all year, he’s sticking with it.
If you asked him when he first made the decision, he’d probably have said something about Elvis being able to ‘make things happen for us’ similar to you suggest, but he’d also say something about him playing hard, giving options, putting energy in the lineup, being comfortable in that spot. Okay, that’s not good reasoning either.

I thought Moreland was traded to the Cardinals before the game started?

I didn’t really watch Pujols on that fly ball, but of course he should be running it out in the freaking WS!  He is one of those guys that almost never runs out balls that are likely to be outs.  I don’t care how good he is or how much money he makes. If I am the manager, he either runs them out or doesn’t play.  If nothing else, it is a poor model for the younger guys.

So why was Napoli batting so low early in the season?  He was an excellent hitter going into the season…

Bring in a lefty to pitch to Descalso?  Oliver?  Tony is not going to pinch hit for Descalso…

Leaving Adams in against Descalso the lefty rather than bringing in Gonzalez… seems like a close decision. I’d probably lean the other way, but it’s very slight.

@13 Because he started doing it and kept doing it. Winning, duh!

More seriously, I doubt it hurts anywhere as much as Andrus batting second, though it’s less explicable. Baseball has a long, stupid tradition of batting crappy hitters second, not of batting your best hitter seventh.

If I am the manager, he either runs them out or doesn’t play.  If nothing else, it is a poor model for the younger guys.

Another point of view is that Pujols manages his exertion to keep him on the field more often. I don’t really need him to pull up lame trying to beat out a routine grounder to short, or seeing how close he can get to 2B on a ball that the OF catches.

He hustles on defense. He steals bases. He takes an extra base on occasion. He picks his exertion times.

If the ball is off the wall, he coasts in for a double regardless. He was at 1B by the time the ball was caught.

I think sometimes we take that poor role model for young players too far. Pujols works his ass off, is extremely charitable off the field, etc.

I would simply tell a kid “He does this 700 times a year, you do it 50 ... now RUN!”

Bunting with Andrus here? This even goes against ‘conventional wisdom’ not to play for a tie on the road…
And as I type this, he stole. Interesting.

I don’t mind a bunt some of the time, because Andrus is so fast and will get lots of ROE and bunt singles.  But with 1-0, you gotta swing away, especially with the infield playing in.

Ballsy steal!  A bunt here would be terrible. 2-1 count!

Never mind, it is 1-2.

Wow, a ballsy steal. Even with a big jump on a closer, you’re running against a quick release, 97mph, and a catcher whose right arm has cybernetic components in it.

That was a helluva throw to make it that close.

Funny, if you are going to have a crappy hitter batting second, it should be a lefty batter - fewer DP and more hits to RF with the leadoff runner on second.

McCarver wants an error on Pujols?  He is out of his mind!

It is a freaking hit and a base advance on the throw.  If there is an error it is on the RF’er for a bad throw!

The folly of batting Andrus 2nd was on display there ... and Wash got lucky.

However, Andrus flashed some heads-up baserunning there. Washington obviously couldn’t have predicted it, but that was nice to see.

Shoulda left Carpenter in. Ha Ha Ha.

Pujols doesn’t deserve an error, but he HAS to cut that ball off. He has to be closer to the mound/center on that one. Slow getting over and it cost a base.

The throw from Jay was on line, which is all you really want from a throw coming from CF.

I let Pujols take it easy in the reg season.  In the post season, he runs everything out!

Good (and ballsy) move by TLR.  Motte has a large platoon split, mainly because fastballs have a large split.

Rhodes is much better than Motte against lefties (near 1 run per 9).  And Hammy has large splits as well.

Of course, you have a bunch of RH hitters coming up, so you have to consider that.

Was anyone else hoping that TLR would bring in Rhodes while putting Motte in LF for one batter?

"The throw from Jay was on line..”

Are you sure?  Watch the video…

I have Lynn as an awful pitcher.  Has he been “hot” for a while or something?

So you take out Motte so that Rhodes can pitch to one batter and then you have to bring in Lynn for 2 or more batters?  I’m not sure anymore that taking Motte out was a good idea.  In fact, I doubt that it was…

Hmm, instant run expectancy of Rangers, bases loaded, no out with Motte still on mound vs. runner on 3rd, one out, and Lance Lynn on the mound?

ZIPS has him as an awful pitcher too.  Don’t know about the other projection guys…

"Was anyone else hoping that TLR would bring in Rhodes while putting Motte in LF for one batter?”

That might have 100% been the correct move…

Taking out Motte for Rhodes… surprising, but probably the right move. Steve Busby saying he would have walked Hamilton there with the righties up after (terrible).

The guy with the best projection in the bullpen (of those left) is actually Jake Westbrook, though he’s a GB guy and not ideal for a man on third w/ 1 out situation. I’d rather have him in than Lynn, though.

Too predictable on the 3-2 pitch to a smart guy like Young.  Braking pitch coming 80% of the time.

McCarver was right about that:  “Young could not have hit that slider without looking for it.”

I would have gone for the high fastball.  If he walks him, it is not a big deal.  He has to avoid contact…

Westbrook is a terrible starter, but as a reliever, maybe he is 1 run better.  Only TLR knows that. As you said, Phil, you gotta choose the guy who has the most K, you don’t mind walks.  You want a guy who misses bats.

You bring in a high K/BB guy, and if Young walks, you bring in Westbrook, the sinker-baller.

Ah, I should have been a manager or pitching coach!

Remember I said, There would probably be a mistake before the game ends.  There were maybe 5 mistakes.

Circle, that throw was terrible. Probably 15 feet off line. I don’t even think that Pujols touched the ball…

When I saw the play, IIRC, Jay was on the 2nd base side of 2B and throwing home. His throw would not “go over the mound” because of his position and where Pujols would be as the cut-off man.

Just saw the replay, Jay was damn near in RF.

The throw was up the 3rd baseline.

I have no idea what Pujols was doing. It’s like he fell asleep for a second there. That’s his cut, AND it’s gloveside.

You don’t ever walk anyone with no outs (other than bottom of the 9th)!

My prediction:  If Cards win, Tony is branded as brilliant again. If they lose, they (media) will second guess him.  Stupid, stupid, stupid…

Was anyone else hoping that TLR would bring in Rhodes while putting Motte in LF for one batter?

... AND hoping for a routine flyball to LF!

Long-toss everyone.

MGL: You just got your Punto moment.

Proper bunting (or not) strategy:

1) Batter must be fast enough and good enough bunter (including whether the pitcher is hard to bunt against) to get more than his fair share of hits and ROE.

Punto?

2) You CANNOT be predictable!  If the defense KNOWS you are going to bunt and plays all the way in, a bunt is NEVER the correct play. Too many forces at second and not enough hits and ROE.

Was there anyone who wondered whether Tony was going to bunt or not?

I’ll second guess Rhodes for throwing an above the belt slider ... especially to a lefty batter with a tender groin and not rotating like he usually does.

That’s an easy pitch to get airborne.

Seriously, throw that ball starting middle ending low and away and there’s a good chance Hammy grounds it to 1B/2B. If it’s to Pujols and Kinsler takes off, he’s dead. If he doesn’t, the it’s 2nd and 3rd with 1-out and you start to contemplate scenarios that could get you out of the inning with no runs scored.

Every year, we see terrible managing in the post season, which is one reason why I wanted these threads - to highlight it.

The BE point for a steal has got to be very low.  50, 60%?  Who is on first again?

Circle does any pitcher intend to throw a high slider unless it is a get me over or maybe back door or front door slider?  I don’t think so…

Is this the most American (or French) color scheme ever in the World Series? Dodgers never played the Red Sox…

@49 Laird. It isn’t quite the same situation, but Earl Weaver liked to steal with one out and his pitcher at bat because the break-even point was so low.

Yup, Tony will be criticized for taking out Motte ONLY because it didn’t work, which is a joke.  If it worked, it would have been lauded as genius…

No pinch runners available for Laird?

I would think that all post-season teams should have a designated pinch runner like Roberts in that Boston series against the Yankees…

Hahah, here comes the criticism of Tony for taking out Motte from Lyons and Karros!

The media is so stupid and predictable…

Circle does any pitcher intend to throw a high slider unless it is a get me over or maybe back door or front door slider?  I don’t think so…

Oh, I’m quite certain that’s not where he was aiming for. No matter what door a pitcher is trying to throw through, he doesn’t want it elevated.

That’s the frustrating part about it though, Rhodes has one job in this whole series ... to throw sliders to Josh Hamilton.

I’ll let Rhodes’s father be consoling and understanding, and let Arthur know that he’s still a good person and that there are people who still love him and all that. I just want my sliders to never start above the belt, let alone end above the belt.

His one job is to throw Josh Hamilton sliders down in the zone.

It’s no all on Rhodes, I’m just not the biggest Arthur Rhodes fan in the world. I’m just being a disappointed Cardinals fan.

MGL/#54

Laird was the pinch runner!

Molina walked, and TLR subbed Laird in for Yadi. According to Fangraphs, Laird does have average+ Bsr numbers. He also has pretty average speed scores. I would think that Lohse at least would be a better option based on Lohse’s time in Cincinnati and the times I’ve seen him since then.

Gotta have someone who can steal a base at a 60% ore more success clip.

Circle, if it was a mistake in location, then isn’t “second guessing” a poor choice of words?  Would you “second guess” someone for making an error or striking out?

Anyway, pitchers throw get me over off speed pitches right in the middle of the zone all the time. Of course, in that situation, you have to strike the guy out…

Wow, I’ve been too hard on Circle today!  He’s like the nicest guy on this blog!

"and the fact that if a guy [Andrus] is a good base stealer, you don’t want to have power guys following him….. “

Well that explains why Napoli is 7th! o_O

More respectful, than nice. There was a point earlier where I was tempted to make a “What’s your &%$#@* problem?” response ... but it’s not my house, and I know where I fit in the hierarchy of sabermetrics.

I thought you figured I was just another jag taking cheap shots and acting like I know more than I do, and that’s rarely the case with me.

When I do speak it’s just out of incredible interest in the subject and it’s the only place (along with FG) where I get to have these discussions.

I’m very grateful that you guys offer this blog where we can read and learn this stuff for free. It’s outstanding.

MGL #47, I think that I would almost never want to bunt against Feliz.  A) He throws very hard.  (often 97-98.) B) He throws high very often.  C) He frequently misses the strike zone.  Since I think all these things were known, (or at least knowable), I think bunting with Punto, (or likely anybody), there was very inferior.

Thanks!  With that attitude, you would never be allowed on BTTF (Bash The ”The Book” Fan”.

BTW, swapping Napoli with Young is more advantageous than swapping Andrus with Beltre.  The former generates an extra 7 runs per season, while the latter only adds .5 run per season.  So maybe having Andrus at #2 is not all that bad.

If we just switch Andrus with Napoli though (G-d forbid we bat a slow power hitter in the 2 hole), and kill 2 birds with one small stone, we get 12 extra runs a game or 1.6 wins a year…

TLR’s getting killed for not having faith in Motte as his closer.  One of the MLB Network analysts even went so far to say that, had TLR stuck with Motte, Motte would feel confident that he is the team’s closer.  Now he’s spending the day confused as to his role on the team.  Because the feelings of a closer trump a team’s opportunity to WIN THE WORLD SERIES.

The results-oriented nature of the media - it’s why managers can’t deviate from the “script.”

One has to wonder if Mitch Williams’ “confidence” was enhanced by being trotted out there continually in the 1993 post-season, despite all evidence that he was cooked.

I personally thought Motte was a better option against Hamilton than Rhodes, because I felt Hamilton’s injury made him less likely to be able to handle Motte’s heat, as opposed to Rhodes’ “heat”. It’s interesting that La Russa had the exact opposite take on it in his post-game comments. On the other hand, I was also impressed with his willingness to yank his closer there (he obviously saw something he didn’t like, beyond just the situation). Most managers, I suspect, would have left the closer in to deal with it.

I think, in the regular season, that argument might have more meaning. There are times when there might be some value to “seeing what this guy’s made of”, or trying to build confidence, so to speak. But that stuff, along with individual emotions, has no room in the playoffs. Players and manager can make up over cigars and champagne.

Motte will say and do the right things.

------------------------------------

I think you can leave or pull Motte there. I don’t have the numbers for each scenario, so take it for what it’s worth. I think it’s less about Motte, than it is the situation of now you have to use 2 relievers instead of one. Rhodes to Hammy, Lynn/RHP to the following batters.

Perhaps TLR was thinking if Rhodes K’s Hamilton (like he did in game 1), then you load the bases and have the GB pitcher (Lynn, 57% GB) try to induce an inning-ending DP and end the damn game up 2-0 in the series.

I think A LOT of managers just leave Motte in because well, he’s the “closer”, and there’s no critricism that comes from leaving him in.

People are over-looking the obvious problem ... there’s runners on 2nd and 3rd with no outs. It’s a ridiculous situation to try and get out of without allowing a run.

People are acting liek relievers get out of those situations scoreless all the time. Yeah, managers wish they did.

One of the MLB Network analysts even went so far to say that, had TLR stuck with Motte, Motte would feel confident that he is the team’s closer.

Yeah, because closing the previous playoff games doesn’t build confidence.

What happens to Motte’s confidence if he alone allows the next 2 fly balls and gives up the lead? Well, at least he knows he’s the “closer”.

Highly improbable situation to get out of without losing the lead, and all the experts are acting like a move or non-move would have guaranteed success.

Can anyone analyze the “no doubles” defense ? Presumably Kinsler flies out on normal positioning, but of course he might adjust his swing. What’s the proper trade-off between allowing extra singles and preventing extra base hits in the ninth inning with a 1 run kead? Of course, the successful steal really made a mockery of the “no doubles” strategy, but we have to believe that was a risky move.

Not that it matters, but I don’t know that Kinsler’s bloop single is caught if the OF is playing at normal depth.  Kinsler has power.

The OF (and the corner IF) always has to vary their depth depending on the relative values of the single and extra base hit (based on the score, inning, and baserunners, etc.). So, as usual, it is not a binary thing.  It is a question of how much and when.  It is not easy to analyze (sabermetrically) because it is hard to know exactly how the number of singles and xtra hits changes as a function of OF depth.

I am not even sure that relative value of the xtra base hit is higher than that of the single, in a one run game and no outs in the 9th.  Tango?

I’m confused.  We (saberists) get criticized all the time for not taking into consideration (including in our models/formulas) things like, “Hammy is hurt and may not be able to turn or catch up to Motte’s fastball.”

Yet, LaRussa, arguably one of the best at seeing and utilizing things like that, takes Motte out.

Ah, yesterday, he was a genius and hero, today, he is a goat…

2nd and 3rd no one out, nobody is a genius. Some guys might get very lucky and escape without giving up any runs ... but that’s probably more luck than genius.

I am not even sure that relative value of the xtra base hit is higher than that of the single, in a one run game and no outs in the 9th.  Tango?

That one is easy enough to figure, and I’ll use this chart:
http://www.tangotiger.net/welist.html

If it’s the top of the 9th, home team up by 1, 0 outs, the WE for the home team is .841.

If visiting batter reaches 1B, it’s .725 for home team.

If he reaches 2B, it’s .644.

Reaching 3B: .559.
HR (tie-ing the game): .500

If he makes an out, it’s .912.

That’s all we really need here.  So, the out has a win impact of .071 wins. For a single (or walk), the change in WE is .116, and for a double, it’s .197.  Triple is .282.  HR is .341.

This is what we have:
-.071: out
+.116: 1B, BB
+.197: 2B
+.282: 3B
+.341: HR

To wOBA-ize this, we add .071 across the board to give us this:
.000: out
.187: 1B, BB
.268: 2B
.353: 3B
.412: HR

Here it gets a bit tricky, but not really.  You can skip this part if you want.  Presume the frequency for the above 5 categories are: 67%, 25%, 5%, 0.5%, 2.5%.  The weighted average of the above (excluding out) is: .219.

Glad you are back.  Divide all the above numbers by .219.  We now get this as the wOBA weights:
.00: out
.85: 1B, BB
1.22: 2B
1.61: 3B
1.88: HR

Let’s merge the 2B/3B together, since it’s hard to defense against one, but not the other:
.00: out
.85: 1B, BB
1.26: 2B, 3B
1.88: HR

There’s not that much difference in terms of weighting for the base hit.  The walk though is a huge killer, as you’d expect: you are up by 1, and the last thing you want to do is put the tie-ing run on base so easily.

Insofar as the no-doubles defense: I don’t really see it in play here.  To the extent that you DO care about it, then you just look at the weights of 1.26 and 0.85.  30 doubles and 100 singles is equivalent to 20 doubles and 115 singles.  So that’s the tradeoff: 10 doubles+triples for 15 singles.  If you do that, you breakeven.

That should read: you get 10 fewer doubles+triples AND 5 fewer outs, for 15 extra singles.

That’s a pretty standard exchange in a typical game-state.

With nobody out on a cold night, I can’t see the advantage of playing a “no doubles” defense, even if the batter does have some power.

How many doubles is the “no doubles” defense really going to prevent? My impression has always been that most doubles shoot the gaps rather than going over anyone’s head, and making the outfielders play extra-deep isn’t going to make the gaps any smaller. And on a cold night, the odds of a double going over an outfielder’s head get even smaller.

And I don’t know whether Motte is really as bad at holding runners as the announcers implied, but if the batter has roughly as many stolen bases as doubles during the season (30 SB, 34 2B, 4 CS) AND the pitcher can’t hold a guy on first, that makes the “no doubles” idea even more dubious.

I really don’t know that they were playing no doubles defense.  McCarver seemed to think so, but he says many dumb things.

On a cold night, the OF is going to be playing closer in general anyway. So a no doubles defense is simply playing a little further back than they would on that same cold night. So the cold really has nothing to do with it other than how it might change the frequency of the various events in general.

You play back to prevent doubles not only over your head but in the gaps.  The further back you are the less likely a ball is going to make it to the wall in the gaps.

Again, I am not advocating playing no doubles defense with no outs and no runners on.  In fact, it looks like it is close one way or another.  Obviously with 1 and especially two outs, you are more likely to play no doubles defense (or 1 and 2 outs with a runner on first).

Tango, you say that the BB is a killer in that situation.  But, as we have discussed before, that is misleading. Some might interpret that as meaning that that you have to throw more strikes, pitch to the middle of the plate, etc. with a 1 run lead and no one on.  Of course every base runner is a “killer”.  But, if with a 2 or more run lead, you absolutely don’t want to walk anyone, then, using my “balance theory”, in every other situation, which leaves only a tied game or a 1-run lead, you don’t can’t the walk, and you have to avoid extra base hits, especially the HR.  Obviously that is much more true with 1 or 2 outs, but as I said, if the walk is a complete disaster, with a 2 or more run lead, relative to a neutral game state, which it is, then a walk in all other situations can’t be too bad…

If people are suggesting that Holliday could have caught that ball I would like to remind them of the differences between matt Holliday and Brett Gardner. Rafael Furcal, on the other hand, almost caught the ball. Ian Kinsler versus a hard thrower ... you play deepish.

What are the odds that the same folks would be critical if Kinsler put one over the head of a straight up or shallow playing defense? Why are they taking away bloop singles, you HAVE TO prevent the extra base hit in that situation ... or so they say.

@#76

I’ll take my chances, as StL, in that situation with a runner trying to steal on Motte and Yadi. That same play (in a different inning) killed Philly in a 1-0 game. Motte may not have a great move, but I’d imagine he’s “quicker” to home than most, so the combination of “Motte’s start to home time” and Yadi’s “home to second” time is IMO above average. Once we found out that Kinsler was on his own, I was impressed with his courage to steal. Big hanging balls, as my buddy would say.

Hamilton’s sac fly was pretty standard. elevated slider pulled deep enough to get the run home. Andrus making it to 3rd was big as it was a pretty damn good throw to 3B.

The low and away pitch Young was able to lift airborne for the 2nd Sac Fly was pretty impressive. That’s definitely a “ground ball” pitch. Obviously, I am not happy with losing, but Kinsler’s steal was bigtime ballsy, Andrus taking 2nd was very heady, and Hammy and Young did what they needed, and Young’s situation was difficult.

TEX took a chance and made some smart plays and won the game. IMO, the game was won by TEX, not lost by StL ... Kinler’s bloop-ass hit not withstanding. He can stick that rabbit foot back up his ... calm down Circle, calm down.

With no baserunner, then a walk by definition is going to be as costly as a single.  Regardless of game-state, a walk is going to be a killer versus a single.

HOWEVER:

Here is the RE for each event, with bases empty, 0 outs, which you can calculate easily enough from here:
http://www.tangotiger.net/re24.html

-.25 out
+.40 1b, bb
+.63 2b
+.89 3b
+1.0 hr

We add +.25 across the board to give us:
+.65 1b, bb
+.88 2b
+1.14 3b
+1.25 hr

The weighted average of the above is 0.74, so we divide by that number to give us:
.88 1b, bb
1.19 2b
1.54 3b
1.70 hr

So, that’s wOBA by the bases empty, 0 out state.

We see it’s not terribly different for that same state, in the top of the 9th, down by 1:
.85: 1B, BB
1.22: 2B
1.61: 3B
1.88: HR

If your perspective is bases empty 0 outs in a random situation as your baseline, then the single and walk have a bit less impact in the 9th inning, the double and triple have a bit more impact and a HR has much more impact.

Therefore, compared to that baseline, you would try to avoid the extra base hit more, esp HR.

It depends what the baseline is for comparison purposes.

***

And my love affair with wOBA continues…

OK, now that you’ve analyzed the tradeoff between singles and doubles/triples, how does that compare to what teamsa atually do? The one-run lead is not much differant than the tie game you’ve given stats for, what trade-off are teams actually making in these situations? Does it vary by manager?

Once we found out that Kinsler was on his own,

I missed where we learned this. Who said it?
If it was Washington, he ALWAYS says what makes his player look good based on the outcome - if it worked it was the player’s call (or occasionally one of his coaches, depending on what ‘it’ is); if it didn’t work, he always takes the blame.

Here’s John Sickel’s on Lance Lynn, in a post two days ago listing the Cards’ 20 best prospects:

6) Lance Lynn, RHP, Grade B: Took to bullpen role with ease, still under the rookie innings limit. What he did in the majors doesn’t look like a fluke to me. I might end up going with a B+.

http://www.minorleagueball.com/2011/10/19/2502022/st-louis-cardinals-top-20-prospects-for-2012

"If your perspective is bases empty 0 outs in a random situation as your baseline, then the single and walk have a bit less impact in the 9th inning, the double and triple have a bit more impact and a HR has much more impact.”

Right.  That is what I said that “a BB is a killer” is misleading.

That wOBA weights comparison is fantastic.  That tells is the that the value of the single is slightly less and the value of the double and triple is slightly more (relative to one another), as compared to on average, therefore, the OF should be playing slightly deeper than “average” (of course there is not “average” state - the OF should constantly be moving in and out based on those relative weights) and the corners should be playing slightly more toward the lines.

Again, it is not a binary decision as many people, including managers and commentators make it out to be.  You constantly move around.  If you want to call an extreme (or even medium) depth “no doubles defense” that is fine by me. In that case, you definitely don’t want to be playing “no doubles” defense in that situation (top of 9, down by 1), according to Tango’s table.  You want to be playing a “slightly less doubles” defense…

Here’s John Sickel’s on Lance Lynn, in a post two days ago listing the Cards’ 20 best prospects:

6) Lance Lynn, RHP, Grade B: Took to bullpen role with ease, still under the rookie innings limit. What he did in the majors doesn’t look like a fluke to me. I might end up going with a B+.

Sickels knows a lot more about prospects than I do (I know nothing other than the numbers), but here are various forecasters on Lynn:

Me: 5.01 ERC (3.90 is the average reliever).
Pecota: 4.34 ERA .6 WAR
ZIPS: 4.50 FIP
Oliver: 4.63 ERA

So, all of the aforementioned forecasters have him projected (most are current projections, I think) pretty terribly for a short reliever (or any reliever for that matter).

Sure, all of those projections have a high degree of uncertainty, as does Sickels rating.

My guess would be he is somewhere in between, but who knows. I was commenting (on him being terrible), based on his projections.

FWIW, I use BA America’s top 100 prospects lists and their draft round and number when I do my projections for young pitchers…

Bringing in Lynn might have been TLR’s best option after Rhodes came in. He’s not a grade-A prospect, but he did pitch fairly well for the Cards in the regular season.

Also, in the aftermath of Matt Moore’s gem in his first start against the Rangers, maybe TLR thought the unknown kid was likely to pitch better than his talent level, at least the first time out. Isn’t there research showing that the pitcher has a relative advantage when he’s unknown to the other team? Not only had the Rangers never seen him, but they probably had much less scouting information on him than Dotel or Mitchell Boggs, the other right handers still in the pen.

And while the sample size is quite small, Lynn’s regular season strikeout rate was much better than Boggs’s, and basically the same as Dotel’s. But Dotel gave up more fly ball outs than grounders, while Lynn was the opposite, suggesting that Dotel might be more likely to give up a sac fly.

Jake Westbrook I suppose might also have been avaiable, and he’s quite a ground-ball pitcher. But he doesn’t get many strikeouts.

Lynn’s minor league numbers were pretty good at AA, but so-so at AAA.

MGL, I’m guessing most of those projections have him forecasted with most of his innings as a starter, as that is what he is/was in the minors.  He has a 2.70 xFIP in 35 IP in the MLB.  You definitely seem to be underrating him.

I posted this on FG:

A back of he envelope calculation:

A pitcher fielding a position (very poorly) is a -50 runs per 150 games fielder.  That is around .009 runs per PA.  You could probably leverage that by putting him at a position that is not likely to get a batted ball.  So, call that a .005 run loss.

If you can have one reliever who is even .5 runs per 9 inning better than another reliever (say, Motte over Lynn), and that is not much (Motte could be 1 run per 9 better than Lynn), that would be a savings of .013 per PA, or around a .03 savings if Motte pitches to 2 or 3 batters rather than Lynn.

So it definitely seems worth it to put your best pitcher in the field for one PA if that pitcher is even .5 runs per 9 better than the alternative and even if he pitches to only one more batter, let alone 2 or 3…