Article: Samuel Deduno showing progress in World Baseball Classic

[Parker Hageman]

You can view the page at http://twinsdaily.com/content.php?r=1451-Samuel-Deduno-showing-progress-in-World-Baseball-Classic

[Shane Wahl]

Some things appear to be off in this article, but I realized the same thing about Deduno. This is a good sign. But it just makes this Correia abomination way worse.

[Parker Hageman]

Something is fudged up. Brock's working on getting some solutions as to why the GIFs are not appearing. Check back tomorrow to see it in full glory.

[h2oface]

new......... and improved?

[glunn]

I am a big fan of Deduno and hope that he has found a way to decrease his walk rate. If that happens, he can be awesome. I am setting my DVR for Thursday's game!

[Teflon]

People need to look a pitchers like Deduno a little more closely and understand that the high walk rate which is such a flashing number in his pitching stats is not as negative an indicator as people would have you think given Deduno's low hit rate.

 

Compare two pitchers with similar WHIPs and the pitcher with the higher walk rate will fare better because a walk only puts the batter on first base and only advances base runners in a force situations a single base while hits will result in more total bases for both batters and runners.

 

Liam Hendriks (1.547) and Sam Deduno (1.544) had nearly identical WHIPs for the Twins last year but Hendriks' was based on 11.2 H/9 and 2.7 BB/9 while Sam's was from 7.9 H/9 and a 6.0 BB/9. Hendriks surrendered 184 total bases to batters versus only 112 for Deduno. This means that for the same number of batters put on base, Deduno gave up 65% fewer total bases. (And this doesn't even account for the extra bases by base runners.) In a similar number of innings pitched (85 for Hendriks, 79 for Deduno) Deduno gave up 21 fewer runs.

 

I won't throw Liam's 1-8 mark in the argument because his run support was horrible (3.4 runs per game.) while Deduno's was decent at 5.1. (and, by the way, Diamond got 6.0. Care to guess what kind of pitcher Diamond will be in 2013 with, say, a run and a half fewer per game?)

[gil4]

Compare two pitchers with similar WHIPs and the pitcher with the higher walk rate will fare better because a walk only puts the batter on first base...

 

It also seems like walk rate is an area where pitchers will take a big leap forward, although it usually happens well before age 29 if it's going to happen. Still, a guy who misses bats and stays healthy could very well be on the verge of being a good pitcher.

[bird]

Great post, Teflon. Thanks!

[old nurse]

People need to look a pitchers like Deduno a little more closely and understand that the high walk rate which is such a flashing number in his pitching stats is not as negative an indicator as people would have you think given Deduno's low hit rate.

 

Compare two pitchers with similar WHIPs and the pitcher with the higher walk rate will fare better because a walk only puts the batter on first base and only advances base runners in a force situations a single base while hits will result in more total bases for both batters and runners.

 

Liam Hendriks (1.547) and Sam Deduno (1.544) had nearly identical WHIPs for the Twins last year but Hendriks' was based on 11.2 H/9 and 2.7 BB/9 while Sam's was from 7.9 H/9 and a 6.0 BB/9. Hendriks surrendered 184 total bases to batters versus only 112 for Deduno. This means that for the same number of batters put on base, Deduno gave up 65% fewer total bases. (And this doesn't even account for the extra bases by base runners.) In a similar number of innings pitched (85 for Hendriks, 79 for Deduno) Deduno gave up 21 fewer runs.

 

I won't throw Liam's 1-8 mark in the argument because his run support was horrible (3.4 runs per game.) while Deduno's was decent at 5.1. (and, by the way, Diamond got 6.0. Care to guess what kind of pitcher Diamond will be in 2013 with, say, a run and a half fewer per game?)

Hendricks has shown an ability to be very good at a level for a longer stretch than 4 innings. See last year in the minors as well as 2010. Deduno has not ever shown that. Hendricks has shown that he can be much better than a 1.5 whip pitcher, Deduno has not.

[Brock Beauchamp]

I believe the phrase of the day is "SWEEP THE LEG, SAMMY".

[Brandon]

If he lowers his walk rate to around 4 per 9 then he could be this generations Dave Stewart. He didn't have much success until his late 20's then out of know where won 20 games a year 4 years in a row. While I won't expect 20 wins a year, Dedunno could put together 3-7 years of 12-15 wins as a starter. I think a pitcher like him makes the rest of the rotation better because of the way he pitches is a different look then the other starters.

[cmathewson]

Looking at the raw data from last year is only partly helpful. I'd like to see the trend line. When he came up, he was walking five, six guys a game and pitching five, six innings a start.

 

He did that for about a month and Anderson got him to make some adjustments. Most notably, he had the catcher sit down the middle and Deduno would throw it at the mit and the ball would move over the corner for a strike. Prior to that, he would try to hit the corner and it would move off the plate. Consequently, he was only throwing his fastball for strikes once every fourth or fifth time. Hence all the 3-0 counts. But late in the year, he was throwing strikes with his fastball and guys started swinging early in the count. When they did, they hit weak grounders. When they didn't he finished them off with the slider.

 

Slowly, the walk rate lowered. By the end of the year, he was walking something like 3/9. Meanwhile, his BA against was under .200. He carried that over into winter ball and dominated. Now he's keeping it going in the WBC. The big test will be tomorrow, when he starts against a USA line-up that could beat most all-star teams. Because he's been away from camp, we've kind of forgotten about him. But with Hendriks and Gibson struggling, he could make this team with a strong showing tomorrow.

[Parker Hageman]

Slowly, the walk rate lowered. By the end of the year, he was walking something like 3/9.

 

This is the perfect example of why K/9 and BB/9 are not good indicators of a pitcher's workload -- always (ALWAYS) use K% and BB% over that.

 

In his first 46 innings, he walked 36 batters in 46 innings -- a 5.8 BB/9 ratio. In his last 33 innings - presumably after these changes from Anderson - we walked 17, an improved 4.6 BB/9, right? HOWEVER, if you look at the overall pool of batters faced, Deduno actually walked more hitters in the latter sample (8.5% in last 33 innings versus 5.6% over first 46)

[jimbo92107]

People need to look a pitchers like Deduno a little more closely and understand that the high walk rate which is such a flashing number in his pitching stats is not as negative an indicator as people would have you think given Deduno's low hit rate.

 

Compare two pitchers with similar WHIPs and the pitcher with the higher walk rate will fare better because a walk only puts the batter on first base and only advances base runners in a force situations a single base while hits will result in more total bases for both batters and runners.

 

Teflon's comment suggests a more general point... Statistical regression is difficult to predict with a pitcher whose success depends on a certain degree of natural unpredictability. Sam Deduno's style indeed hinges on peppering the area in and around the strike zone with sliding, ducking fastballs. The downside of such a style is a high walk rate; the upside is a lot of whiffs and really lame contact.

 

One thing I've noticed about Deduno, and the reason I find him fascinating, is that he appears to have developed a style that fits his abilities. He's an excellent fielder, quick as a good shortstop. His whippy arm motion is unusual for an overhand delivery, but it appears not to damage his shoulder or elbow. Ever notice that his throws to first and second base are straight as an arrow and pinpoint accurate? Deduno knows very well how to throw a ball straight.

 

Whether his new, more whirly deliver results in better accuracy is yet to be seen, but it's not impossible. The fact that he falls off so far to his left indicates that he is accelerating in that direction from the start. Does that delivery really look more repeatable to you than his supposedly more disciplined finish from last year? Throwing a baseball combines straight line and rotational movements. When you find a good recipe of such whippy movements, you can wind up with a Roger Federer forehand, or a Sam Deduno fastball for a strike.

[Brock Beauchamp]

This is the perfect example of why K/9 and BB/9 are not good indicators of a pitcher's workload -- always (ALWAYS) use K% and BB% over that.

 

In his first 46 innings, he walked 36 batters in 46 innings -- a 5.8 BB/9 ratio. In his last 33 innings - presumably after these changes from Anderson - we walked 17, an improved 4.6 BB/9, right? HOWEVER, if you look at the overall pool of batters faced, Deduno actually walked more hitters in the latter sample (8.5% in last 33 innings versus 5.6% over first 46)

 

Excellent point, Parker. I tend to fall back in K/9 and BB/9, even though I know it's wrong. I need to stop doing that.

[Mike Sixel]

It would sure help if he could be a marginally decent starter for this team....

[cmathewson]

This is the perfect example of why K/9 and BB/9 are not good indicators of a pitcher's workload -- always (ALWAYS) use K% and BB% over that.

 

In his first 46 innings, he walked 36 batters in 46 innings -- a 5.8 BB/9 ratio. In his last 33 innings - presumably after these changes from Anderson - we walked 17, an improved 4.6 BB/9, right? HOWEVER, if you look at the overall pool of batters faced, Deduno actually walked more hitters in the latter sample (8.5% in last 33 innings versus 5.6% over first 46)

 

I must confess I don't get this. You seem to be saying that his walk rate went up even though his walks-per-innings-pitched went down. As far as I can tell, the only way to make sense of that statement is to say his walk rate went down in a smaller sample of innings than his previously higher walk rate. I don't see how that's helpful. Our judgement of him should not be deprecated because the season ended before he had a chance to put up an even number of innings after the adjustment as before.

 

Either way, we are dealing with small samples, and it's really hard to project such an unpredictable player with such small samples. All I know is, since the adjustment, and including his winter ball and WBC innings, his walk rate is in the area where he can be successful with such a low BA against. Relative to the current competition (Cole DeVries, who walks hardly anyone but gives up a lot of hits) his WHIP should be lower, giving him a better chance at success.

[Brock Beauchamp]

I must confess I don't get this. You seem to be saying that his walk rate went up even though his walks-per-innings-pitched went down. As far as I can tell, the only way to make sense of that statement is to say his walk rate went down in a smaller sample of innings than his previously higher walk rate. I don't see how that's helpful. Our judgement of him should not be deprecated because the season ended before he had a chance to put up an even number of innings after the adjustment as before.

 

It's pretty simple. A guy doesn't face the same amount of batters in every inning.

 

If he K's 5 guys in a nine inning game, that's a 5 K/9 ratio.

 

But let's say he pitched two nine inning games and struck out 5 guys in both.

 

But in one game, he walked 8 batters, gave up 10 hits, and hit 3 batters. That means he struck out 5 of a possible 48 batters (27 outs, 8 walks, 10 hits, 3 HBP).

 

In the other game, he walked 4 batters, gave up 5 hits, and hit no one. That means he struck out 5 of a possible 36 batters (27 outs, 4 walks, 5 hits).

[Parker Hageman]

Here's a more detailed explanation as to why to stop using K/9 & BB/9:

 

What happens when we use the 'per 9' metrics is that we lose accuracy, because our measurements have become subject to the tyrannical forces of BAbip. As a pitcher allows more Hits per Ball in Play, he becomes less efficient. He ends up facing more batters and getting fewer outs, which consequently means fewer innings. But if he's still striking out batters at the same rate (say 20%) all the while, his K/9 is going to look a lot shinier with those fewer Innings.

 

Stop using K/9 and BB/9! - Beyond the Box Score

[Parker Hageman]

Well, Brock laid it out in an easier to digest format...

[Brock Beauchamp]

Well, Brock laid it out in an easier to digest format...

 

i r understend teh simpel things

[kab21]

Great post, Teflon. Thanks!

 

One part that Teflon missed in his analysis was that Deduno's low hit rate was due to a low BAPIP (.266). Hendriks OTOH posted a .313 BAPIP. If Deduno's BAPIP is normal for pitchers then his hit rate will be a lot higher (WHIP also) and those walks are going to hurt even more than they did last year.

 

If he's going to walk a lot of batters then he's going to need to K a lot to be successful.

[Shane Wahl]

I keep getting sucked into my hatred for the Correia signing on a weekly basis around here. Part of that is that I read MLB's Twins offseason analysis and it was just so damn blunt about how terrible the signing was. It isn't just that money is wasted, it is that Deduno and/or DeVries might actually have *better* seasons than Correia if given the chance. And at 1/9 the cost this year and 1/10 of it next year.

[cmathewson]

Point is, these sample sizes make the distinction tenuous at best. I get it over the course of a whole season or something. You're just controlling more variables. But over 80 innings, it's probably not all that helpful. If he's getting more groundouts and fewer Ks over 33 innings than he did over the previous 46 innings, what does that mean? It could mean a lot of things. Here I think Pitch F/X data might be more helpful.

[Parker Hageman]

If he's getting more groundouts and fewer Ks over 33 innings than he did over the previous 46 innings, what does that mean?

 

Well, simply put, it means that Deduno was walking MORE batters in his second-half of the season after Anderson made some mechanical changes not less. Big picture, using innings as the denominator for measurement of strikeouts and walks paints an inaccurate picture of a pitcher's success versus what their true workload is: batters faced.

 

One part that Teflon missed in his analysis was that Deduno's low hit rate was due to a low BAPIP (.266). Hendriks OTOH posted a .313 BAPIP. If Deduno's BAPIP is normal for pitchers then his hit rate will be a lot higher (WHIP also) and those walks are going to hurt even more than they did last year.

 

There is a significant difference between how opponents were putting the ball in play against Deduno compared to Hendriks which led to that vast chasm in BABIP marks. Hitters were not squaring up on Deduno. For example, Deduno had a 58% groundball rate versus Hendriks' 41%. Meanwhile, 17% of the fly balls hit off of Deduno did not leave the infield. More weak contact. Both are likely to equalize closer to league average over the course of a regular season but I would wager Deduno would be able to keep his beneath the average while Hendriks would stay over that figure.

[Nick Nelson]

Teflon made a great point in his earlier post, and it's something I spoke to last season. Deduno is what you'd call "effectively wild"; a high walk-rate is part of his game. He doesn't know where his pitches are going to end up and neither do opposing batters, contributing to consitently weak contact evidenced by low hit rates and homer rates over the course of his career. It's true that you'd rather have a guy walk to first than run to second or third, or jog around the diamond.

 

That being said, a BB/9 of 6.0 is unpalatable, so Parker's writeup here is encouraging.

[YourHouseIsMyHouse]

I really like the gifs in this. The movement on the pitch (despite the uglier leg kick) was really awesome. Also, that was an amazing squeeze in the 2nd.

[kab21]

There is a significant difference between how opponents were putting the ball in play against Deduno compared to Hendriks which led to that vast chasm in BABIP marks. Hitters were not squaring up on Deduno. For example, Deduno had a 58% groundball rate versus Hendriks' 41%. Meanwhile, 17% of the fly balls hit off of Deduno did not leave the infield. More weak contact. Both are likely to equalize closer to league average over the course of a regular season but I would wager Deduno would be able to keep his beneath the average while Hendriks would stay over that figure.

 

It should be worth noting that GB's have a higher BAPIP than FB's. They are of course way better than FB's since they only go for singles. it's the line drives that kill pitchers and neither Deduno nor hendriks did well there.

 

Bottom line though is that Deduno needs to approach a 2:1 k:bb ratio to be halfway successful imo. I don't care how he does it (more K's or less BB's) but a 1:1 k:BB ratio belongs in AAA.

[Oldgoat_MN]

I have long felt that a little wildness is not always a bad thing.

Deduno is a wildcard, but I hope he gets a shot as a SP.

It would be wonderful if he could surprise us. (and all of MLB)

[bear333]

Some things appear to be off in this article, but I realized the same thing about Deduno. This is a good sign. But it just makes this Correia abomination way worse.

 

I'm concerned about Correia, as well. By the way, there were only two Twins starters with a winning record last year and Deduno was one of them. Even with his walks he may end up being more effective then a couple of the guys we are going to start the season with. I would like to see him as a 5th or 6th starter, to be honest. I love his wicked fast ball.