I wake up out of a dead sleep, screaming, drenched in sweat… "God, not another GIDP."

Josh Reddick said yesterday "The team hitting into six double plays isn’t frustrating if you win"

Okay, fine Josh, I’m happy for the win, but I don’t know if I would say it isn’t frustrating. This year, the Astros have been plagued by them. Last night, we set a new all-time franchise record by grounding into 6 double plays in a game. It’s a consistent gut punch, especially with the offensive sputtering recently. With us leading all of the majors in this terrible stat, I wanted to do a deep dive into it.

According to Fangraphs, a double play costs the team almost a full run, or about three times as much as an ordinary out.

Initial Theories / OBP

So to start with, there has been a lot of comments that it is the nature of the beast with having a high OBP team that makes a lot of contact. As of today, the Astros are tied for 7th in OBP. Of the top 7 teams in OBP, we are the only one who is is in the top 7 of GIDP, so that theory doesn’t seem to have a strong correlation.

So I looked to advanced stats, and discovered a new one wGDP. There is a good write up on fangraphs on the stat here ( https://www.fangraphs.com/library/offense/wgdp/) but in short, it basically weighs players based on how good they are at avoiding double plays vs the number of opportunities they had (less than 2 outs, runner on first base). The rating system from their site is as follows:

I’m sure you can guess, but the Astros came in at second worst in all of baseball (Thanks Orioles for being worse than us!) At -4.0 as a team. Considering -2.5 is Awful, we’re off the charts bad.

But as for our players, I figure there are a few essential components of a GIDP, which are common sense. We covered the number of opportunities, and Fangraphs somewhat shows that we’re excessively terrible and the correlation is not strong to the OBP = GIDP theory. So I’m going to analyze ground ball % (hard to hit into a GIDP if not on the ground), Speed (figuring if you are slower, you’re less likely to beat out the second out), and hard hit ball % (faster it gets to the fielder, the more time they have). I’ll compare all of this to last year (where we were 21st best in wGDP and came in 6th with 139 GDP.

So there is a TON of data here, don’t worry, it’s more of a reference point, and I’ll dive into the data to try to find the biggest causation.

I sorted this out to players with more than 50 PA, just to avoid pitchers and some others disrupting data. All of these results are just used averaging the players results, not calculating out by number of attempts * results then averaging – not a perfect analysis but I believe it’ll be good enough to hold the discussion.

Ground Balls

This seems like the most obvious place to start, since it’s tough to GIDP without hitting it on the ground.

In 2018, the league average is 43.3% of all balls hit are hit on the ground. This is actually the lowest % since 2002, but it has remained relatively consistent (43.3% - 45.3%).

The Astros come in at 17th out of 30 (44.1%), so while worse than league average, there wasn’t as strong of a correlation as I expected.

Additionally, none of the teams in the top 5 in GB %, showed up in the top 5 of GIDP

Speed

There are a few stats that are used to calculate speed and base running ability..

Spd

First up is Spd, it’s a bit of an outdated Bill James stat, which has mainly been replaced by UBR, but it weighs SB, frequency of attempts, runs scored, percentages of triples etc

By SPD, we come in at an average of 3.94. This would rank as below average (4.4 is average this year). This stat does seem to match the "eye test" of who I would normally consider fast.

Sprint Speed

Sprint Speed by Statcast is "feet per second in a players fastest one-second window". It does remove routine ground outs. League average is 27 feet per second.

Averaging all of our players, we come in at 27.04, so basically exactly average on speed We have 9 players that rated as above average, 7 that rated as below average.

UBR

Ultimate Base Running works to create a weighted runs gained / lost based on a players base running. It basically determines how often the runner advances a position, or gets thrown out and compares to the odds of how often it would normally happen. The Astros have been known to be aggressive in their running and seemingly TOOTBLAN regulars.

Interestingly the Astros come in at a team at 1.5 UBR against a league average of 0.7. I think this lends some credibility to the aggressive approach based on analytics.

Average Exit Velocity

Again, nothing ground breaking here, and obviously I didn’t have the data for just ground ball EV, but the hitters above averaged out to 88.11 average EV.

Summary

So this was a bit more frustrating of an exercise than I expected. I was hoping for the "Magic bullet" or cause be blaringly obvious. I guess if it was Luhnow would have fixed it already. It seems like a combination of a ton of opportunities, with slightly below average speed, hitting more ground balls, and hitting them hard, seems to have made a perfect storm.

Others Research

I tried to do some research on it, and interestingly enough, found the Astros leading the league back in 2009, and an old TCB article (https://www.crawfishboxes.com/2009/9/11/1025837/the-astros-and-grounding-into) trying to find the exact same causation, with similarly frustrating results. Cool to see some of the same commenters back then.

I dug in and found an old Fangraphs article (https://www.fangraphs.com/tht/the-worst-thing-a-batter-can-do/) that found a strong correlation between batters behind right vs left handed (Of the 20 worst GIDP culprits 19 were righties, of the 20 best at avoiding it 19 were lefties). This does seem to correlate somewhat to our team and to the wGDP stat. But that and his other article on the same subject didn’t seem to come to any other big conclusion to find causation.

The other noticeable trend I found was that a higher K% often resulted in a lower wGDP score. This makes sense, but doesn’t give a lot of indication.

So based on this, we need more super fast, high K, high Slugging, low GB hitting Left Handers.

Curtis Granderson has the best wGDP of anyone in baseball from 2010-2018 – and is available after clearing waivers (I’m not seriously advocating trading for him)

I think tomorrow I’ll do a second part to this, I found statcast has some interesting components in regards to situational pitches, so I will try to see if I can identify any notable trend from the league’s approach against us instead of our batters.

Let me know your thoughts, missed methodology or stats.

None the less, we WON! Holding onto that 1st place spot in the AL West!!!