2009 LIPS ERA Leaders

Hey guys,
David Gassko informed me that he’s recently made some tweaks to the LIPS formula, so here are updated numbers to account for them. The list is pretty similar with a few changes.

LIPS ERA Top 25 (through 6/2/09)

+------------+----------+----+------+------+----------+-------+------+-------+--------+
| LAST       | FIRST    | GS | IP   | ERA  | LIPS ERA | K/9   | BB/9 | xGB%  | IF FB% |
+------------+----------+----+------+------+----------+-------+------+-------+--------+
| Greinke    | Zack Z   | 11 | 82.0 | 1.10 |     2.77 |  9.66 | 1.32 | 44.33 |   3.94 |
| Santana    | Johan    | 11 | 72.0 | 2.00 |     2.83 | 11.13 | 2.63 | 32.79 |  11.48 |
| Lincecum   | Tim      | 11 | 71.7 | 3.01 |     3.13 | 11.43 | 2.64 | 45.90 |   2.73 |
| Vazquez    | Javier   | 11 | 70.3 | 3.58 |     3.16 | 11.00 | 2.05 | 45.30 |   3.87 |
| Peavy      | Jake     | 12 | 74.7 | 4.10 |     3.20 | 10.13 | 3.13 | 40.64 |   4.28 |
| Halladay   | Roy      | 12 | 91.0 | 2.77 |     3.26 |  8.11 | 1.19 | 55.94 |   2.68 |
| Haren      | Dan      | 11 | 78.0 | 2.42 |     3.30 |  9.00 | 1.15 | 40.39 |   2.96 |
| Verlander  | Justin B | 11 | 69.3 | 3.63 |     3.36 | 11.68 | 2.60 | 31.58 |   2.92 |
| Harden     | Rich     |  8 | 43.7 | 4.74 |     3.37 | 10.92 | 4.33 | 38.53 |   9.17 |
| Weaver     | Jered D  | 10 | 68.7 | 2.36 |     3.40 |  6.95 | 2.36 | 33.16 |   7.65 |
| Hamels     | Cole     |  9 | 48.3 | 5.21 |     3.42 |  9.31 | 1.86 | 43.48 |   2.90 |
| Johnson    | Randy    | 10 | 52.0 | 5.71 |     3.63 |  9.35 | 3.29 | 46.21 |   3.45 |
| Hernandez  | Felix A  | 11 | 71.3 | 3.41 |     3.67 |  9.08 | 2.52 | 49.76 |   2.44 |
| Johnson    | Josh     | 11 | 74.3 | 2.66 |     3.68 |  7.75 | 2.18 | 54.98 |   2.84 |
| Slowey     | Kevin    | 11 | 68.0 | 3.97 |     3.70 |  6.49 | 0.93 | 34.63 |   5.63 |
| Jackson    | Edwin    | 11 | 74.3 | 2.30 |     3.74 |  6.90 | 2.18 | 36.28 |   6.05 |
| Scherzer   | Max M    | 10 | 54.3 | 4.47 |     3.85 |  9.44 | 3.64 | 43.14 |   2.61 |
| Oswalt     | Roy      | 12 | 69.3 | 4.28 |     3.91 |  7.14 | 2.47 | 41.23 |   5.69 |
| Lester     | Jon T    | 11 | 65.3 | 5.65 |     3.92 | 10.19 | 3.31 | 45.16 |   2.15 |
| Bedard     | Erik     | 10 | 60.7 | 2.37 |     3.92 |  9.05 | 2.67 | 41.72 |   4.29 |
| Baker      | Scott S  |  9 | 52.7 | 6.32 |     3.94 |  6.66 | 1.71 | 29.07 |   8.14 |
| de la Rosa | Jorge A  | 10 | 54.7 | 5.43 |     3.97 |  9.38 | 4.28 | 43.92 |   4.05 |
| Gallardo   | Yovani   | 10 | 65.0 | 3.18 |     3.98 |  9.00 | 3.32 | 43.68 |   2.30 |
| Richmond   | Scott    |  9 | 54.0 | 3.50 |     3.99 |  7.33 | 3.00 | 41.25 |   4.38 |
| Pavano     | Carl     | 11 | 63.0 | 5.29 |     4.01 |  7.14 | 1.86 | 46.23 |   3.52 |
+------------+----------+----+------+------+----------+-------+------+-------+--------+

Also, there was a good comment that I think should be given more attention: “Does this mean I should pickup and stash Blanton and De La Rosa?” The answer is “No, not necessarily.” What LIPS ERA gives us is a luck-neutral indication of how well a pitcher has performed so far this year. It is a much more solid indicator than ERA, but it is not the be-all-end-all. Just because Edwin Jackson has a 3.74 LIPS ERA right now does not mean he will post a 3.74 ERA going forward.

What we’re looking at right now is a 65 (or so) inning sample of a player’s true pitching ability. This sample is relatively small in the grand scheme of things and should not be the only thing considered. To better estimate a pitcher’s true ability, we need to look at a larger sample — i.e. his performance in previous years. While LIPS ERA is much more stable than actual ERA, it is still prone to small sample size caveats.

As an example, let’s say you go to a restaurant and have an awful meal. While the restaurant may truly be an awful one, we can’t say for sure after one single meal. Maybe you go there another five times and have really good meals. The more times we go to the restaurant, the more accurate we will be when we talk about the overall quality of the restaurant. If we simply judged it by any single meal, though, the chances of being wrong would be relatively large.

If we roll a six-sided die twice and it lands on ‘3’ both times, are we going to say that this die is more likely to land on ‘3’ than any other number? Of course not. The sample we’re basing this on is too small. If we roll that die another thousand times, I can assure you it will land on each number about evenly.

The same logic applies here. Right now in 2009, we’re looking at one meal (or one die roll). For the guys on the list, it’s a very good meal, but one meal nonetheless. It’s certainly better than if it were a bad meal (or a bad LIPS ERA), but we still need more to go on before we make any definitive assertions. A lot can happen over 65 innings. For those who followed CAPS in the off-season, we see that a lot can happen with a pitcher’s peripherals over the course of an entire season.

One thing to keep in mind is for leagues where you can stash players on your bench. In this case, even though Carl Pavano may not be a true 4.01 ERA pitcher, it might be worth holding onto him to see if he is. Maybe his true talent level has changed and we just don’t know it yet because the ‘meals’ that will tell us this haven’t happened yet — they’ll be happening throughout the rest of the season.

Hopefully this helps put things into better perspective for everyone.


21 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
bpasinko
14 years ago

Does this mean I should pickup and stash Blanton and De La Rosa?

Marco Fujimoto
14 years ago

bpasinko,
In short, yes, this table seems to suggest that those two are worth stashing. Of course, the transactions depend on various aspects like league format, player to be dropped, etc.

What this table essentially says is that Blanton and De La Rosa have both pitched better than what their current ERA’s seems to indicate. And in essence, a couple of the goals behind LIPS ERA are to paint clearer pictures of pitcher performances and to give us an idea of what we should expect in the future.

bpasinko
14 years ago

Blanton’s a head scratcher.  His FIP is almost 5 and his xFIP is over 5, yet his LIPS is a great 3.82? 

De La Rosa has fine numbers in every advanced ERA metric so that doesn’t surprise me, but Blanton does.

Wade
14 years ago

Need translations for: LIPS, xGB%, and IF FB%.  If someone’s gotta second to post ‘em, I’d like to be able to understand the scoring better.

Thanks.

Have a good Wednesday.

Andrew
14 years ago

Thanks, Derek.

It’s truly impressive what you’re doing in NL LABR. I hope the fantasy community is beginning to realize that you’re the best in the business. I didn’t think there would ever be anyone like Ron Shandler, but you rank right up there with him for my money.

Derek Carty
14 years ago

First, the list has been updated based on some recent tweaks David Gassko made.

bpasinko, I answered your first question in the article.  Very good question.

Blanton has fallen off the list now, and his LIPS ERA is now 4.16.  His FIP is 4.90 because FIP doesn’t account for HR/FB, which most pitchers have little control over.  Blanton’s is a very unlucky 16%, inflating his FIP.  FIP is actually one of my least favorite ERA estimators (along with ERC).  His xFIP is 4.09 (http://www.hardballtimes.com/thtstats/main/player/4849/); I’m not sure where you’re seeing that it’s over 5.00.  If we want to look at even more ERA estimators (even though I prefer LIPS), QERA says 4.16 and xERA says 4.10.  So while LIPS was originally a little more optimistic, it wasn’t too far off, and now it’s right in line.

Derek Carty
14 years ago

Wade,
LIPS stands for Luck Independent Pitching Statistics.  It was invented by David Gassko here: http://www.hardballtimes.com/main/article/dips-lips-and-hips/

It essentially tells us what a pitcher’s ERA would be with neutral luck.  There are other estimators that do this (FIP, xFIP, xERA, etc) but I think the LIPS methodology is the soundest and it has shown to be quite predictive.

xGB% (Expected GB%) is a pitcher’s GB% given a league average LD%.  LD% is something a pitcher has little control over, so pitchers with abnormally high or low LD% won’t be judged properly but looking at unaltered GB%.

IF FB% is Infield Fly Ball Percentage.  LIPS is the only ERA estimator that I know of to include IF FB%, but it’s an important factor since 98% of infield flies become outs.  It’s essentially like extra strikeouts.

Derek Carty
14 years ago

Andrew,
Thanks very much for the kind words.  I’m certainly very pleased with my performance so far in LABR, and it’s nice to get noticed.

Thanks!

bpasinko
14 years ago

Derek, thanks again.  My mistake on Blanton, I quickly glanced and looked at his xFIP from 2008 not this year.  That makes it much less confusing.  I suppose you could do a lot worse than a spot start for Blanton, especially away.

wilfork for food
14 years ago

Just traded aaron hill and zambrano for vasquez based on your statistical analysis.  Tell me that’s a steal please

Mike
14 years ago

Thanks Dave,

Follow-up question

Do batters control their IF FB%?

Should I be adding IF FB% to K% when looking at the number of outs a batter is directly responsible for? Or should I assume IF FB% will regress?

Mike

David Gassko
14 years ago

Yeah, you can do that.

Toffer Peak
14 years ago

“Also, there was a good comment that I think should be given more attention: “Does this mean I should pickup and stash Blanton and De La Rosa?”“

I think another thing that people need to consider is that these ERA estimators are used to evaluate pitchers assuming neutral luck and park effects; that makes sense if you’re a GM looking to acquire a real life pitcher. But for fantasy owners, they must keep in consideration that a pitcher’s bad luck may not even out if they player for a team with a bad defense, that will inflate BABIP, or in a hitter’s park that will inflate HR/FB. That’s why I would love if THT or another start would provide a fantasy LIPS or xFIP that accounts for the player’s real life team’s BABIP, HR/FB, etc.

“LIPS is the only ERA estimator that I know of to include IF FB%, but it’s an important factor since 98% of infield flies become outs.  It’s essentially like extra strikeouts.”

tRA also accounts for IF FB%. http://statcorner.com/tRAabout.html

Toffer Peak
14 years ago

Also, is there any reason THT doesn’t include LIPS ERA in the Stats pages?

Derek Carty
14 years ago

Mike,
As David said, pitchers can control their IF FB%.

As far as adjusting FIP, if left with no alternatives, I’d simply eyeball the pitcher’s HR/FB and if it’s too high or too low, the FIP should be higher/lower accordingly.  This brings a lot of guesswork into, though, when it’s easier just to check out something like xFIP or tRA which does all the work for you.

bpasinko
14 years ago

That’s darn impressive by Johan.  He’s gotta have the biggest percentage of at bats that turn into Ks or IFFB right?  Such a large percentage of his outs come by strikeout or a play that even I could make on the infield.  It looks like Harden is close, but does anyone rely less on defense to be effective than Johan?

Mike
14 years ago

Derek,

Two questions

1-Do pitchers have control over IF FB%? I was under the impression that IF FB% was more like HR/FB or LD% so I assume it will regress over time.

2-I use another site for sortable statistics (wont put the link in here). FIP is the luck neutral statistic they provide. Is there a quick rule I can apply to adjust FIP to be more meaningful. I’ve just been eyeballing is and, if the HR/FB% is abnormal, making mental note of it.

Thanks.

Toffer Peak
14 years ago

Mike asked, “Do pitchers have control over IF FB%? I was under the impression that IF FB% was more like HR/FB or LD% so I assume it will regress over time.”

As David and Derek said above I believe pitchers can control IF FB%. In fact, running a regression on GB% seems to show a pretty decent correlation with a .35 R^2.

Just looking at some of the leaders and laggards of IF FB% seems to show that flyball pitchers are quite good at getting infield flys and groundball pitchers are quite poor.

Webb, Maddux, Lowe and Carmona, four of the most extreme groundball pitchers are the in the bottom five for IF FB% over the last two years. It makes sense, IF FB% require the ball to get pretty far below the ball and when you’re throwing most of your pitches low in the zone (as GB pitchers do) it’s going to be pretty difficult to do so. Like wise, most of the leaders in IF FB% are extreme FB pitchers since they’re mostly trying to beat you with high heat (Johan, Peavy, Hamels, Ol. Perez, etc).

Toffer Peak
14 years ago

David said, “LIPS actually accounts for a player’s ballpark, so that it can be directly compared to ERA. It’s true that it does not include a team’s fielding prowess, but teams do not differ THAT much in their fielding ability, and LIPS is all about finding big differences between what a player’s ERA is and what it should be—you shouldn’t be using it to make decisions when the differences are small, and when they’re big, it doesn’t matter what a team’s fielding ability is.”

So it sounds like LIPS is a fantasy tool and not for people to compare players in real life? If so I like that.

I disagree that fielding is unimportant though. UZR suggests that teams can vary by as more than 120 runs over a season. For a pitcher going 200 innings that can mean a difference of 16 runs. If park factors are important to LIPS I don’t see how fielding can’t be.

Either way, I’m now more interested in LIPS and may be willing to look into paying for Heater (I already enjoyed THT’s pre-season projections) as Baseball Prospectus is really losing its value to me.

Andrew
14 years ago

Surprised not to see the following SP’s:

Sabathia
Billingsley
Wainwright
Shields
W Rodriguez
Lowe

David Gassko
14 years ago

Toffer Peak,

LIPS actually accounts for a player’s ballpark, so that it can be directly compared to ERA. It’s true that it does not include a team’s fielding prowess, but teams do not differ THAT much in their fielding ability, and LIPS is all about finding big differences between what a player’s ERA is and what it should be—you shouldn’t be using it to make decisions when the differences are small, and when they’re big, it doesn’t matter what a team’s fielding ability is.

LIPS isn’t included on our stat pages because it is very difficult to calculate, especially in a wholly automated way. John Burnson, however, should be including it in Heater Magazine every week starting with the next issue. Read more about Heater here: http://www.hardballtimes.com/main/downloads/heater_intro

—-

Mike,

Pitchers definitely control their IF Fly%. It’s pretty stable from year to year.