Matt Belisle's 2017 Season

[Seth Stohs]

So, overall this year, Matt Belisle has pitched 45.2 innings in 47 games. He's got a 4.34 ERA and a 1.27 WHIP.

 

Now, he's had four HORRIBLE games:

 

1. April 26 - 0.2 IP, 5 ER, 3 H, 2 BB
2. May 7 - 0.1 IP, 6 ER, 4 H, 2 BB
3. May 29 - 0.1 IP, 3 ER, 2 H, 1 BB
4. June 11 - 0.2 IP, 4 ER, 4 H, 1 BB

 

So, we all know that you can't just pick and choose which games to count in numbers, but just for fun, I thought I'd do it anyway.

 

So, in the 43 games that Belisle has pitched in (not counting the 4 above), he has thrown 43.2 innings. He has a 0.82 ERA and a 0.89 WHIP! That's pretty crazy!

 

 

Those four games... 2 IP, 81.00 ERA, 8.50 WHIP.  Also, pretty crazy!

 

He hasn't given up an earned run since June 26 (16 outings), and a run period since June 27 (14 outings). An inherited runner hasn't scored off of him since July 1st. Only 4 of 23  (17%) of inherited runners have scored on him all season.

 

In other words, he's been great this year... except for those four games... (and just to repeat, I'm not saying we can just eliminate the bad stuff, but perspective is good too)

 

 

[Vanimal46]

He's been sneaky good for 1.5 months... Glad to see him turn things around after a very rough start to the season. 

[Cory Engelhardt]

Love this, thanks for sharing!

[Tom Froemming]

I stumbled across this yesterday ...

 

Matt Belisle 2017

0 days rest: 8.22 ERA

1 day rest: 7.63 ERA

2+ days rest: 0.79 ERA

 

I think he'll continue to be very reliable as long as Molitor doesn't burn him out. 

[Brandon Warne]

He's been great since San Francisco. He also smiles like Gary Busey.

[Willihammer]

 

I stumbled across this yesterday ...

 

Matt Belisle 2017

0 days rest: 8.22 ERA

1 day rest: 7.63 ERA

2+ days rest: 0.79 ERA

 

I think he'll continue to be very reliable as long as Molitor doesn't burn him out. 

That's exactly why he needs another couple arms he can trust IMO. Can't just be the Rogers-Belisle show when they inevitably hit a stretch of close games.

[Vanimal46]

 

That's exactly why he needs another couple arms he can trust IMO. Can't just be the Rogers-Belisle show when they inevitably hit a stretch of close games.

 

Agreed. They should probably still be on the lookout for another rental RP arm. However, Hildenberger is looking promising for a higher leverage role. 

[Seth Stohs]

 

I stumbled across this yesterday ...

 

Matt Belisle 2017

0 days rest: 8.22 ERA

1 day rest: 7.63 ERA

2+ days rest: 0.79 ERA

 

I think he'll continue to be very reliable as long as Molitor doesn't burn him out. 

 

 

April 26 - 0.2 IP, 5 ER, 3 H, 2 BB (1 day rest)
May 7 - 0.1 IP, 6 ER, 4 H, 2 BB (1 day rest)
May 29 - 0.1 IP, 3 ER, 2 H, 1 BB (0 days rest)
June 11 - 0.2 IP, 4 ER, 4 H, 1 BB (0 days rest)

 

So again, those 0 days rest and 1 days rest prove to be pretty meaningless because they're so drastically affected by 2 outings. He has pitched in 8 games on 0 days rest, and 18 games on 1 day rest... Those numbers can be majorly effected by one or two bad outings. 

[Willihammer]

Seth if we're going to throw out the appearances where a reliever gets knocked around then literally every reliever in baseball is awesome.

[Willihammer]

He's 37, its not a stretch to think he'd perform significantly worse, on average, without adequate rest.

[wsnydes]

 

Seth if we're going to throw out the appearances where a reliever gets knocked around then literally every reliever in baseball is awesome.

So you're saying it IS easy to find relievers!  

[Mike Sixel]

This is why I don't like mushing season stats together and coming up with averages for pitchers all that much. Sure, there is value there, but it misses a really important thing, imo.

 

Games are discrete events. I'll take the relief pitcher that gives up 10 runs in one game, and zero for the next 4, over the guy that gives up 2 runs in three games, and 1 run in the next two games (almost) any day of the year. I am NOT suggesting throwing out the bad games, but I think stats like what 538 did make more sense.

[Tom Froemming]

 

April 26 - 0.2 IP, 5 ER, 3 H, 2 BB (1 day rest)
May 7 - 0.1 IP, 6 ER, 4 H, 2 BB (1 day rest)
May 29 - 0.1 IP, 3 ER, 2 H, 1 BB (0 days rest)
June 11 - 0.2 IP, 4 ER, 4 H, 1 BB (0 days rest)

 

So again, those 0 days rest and 1 days rest prove to be pretty meaningless because they're so drastically affected by 2 outings. He has pitched in 8 games on 0 days rest, and 18 games on 1 day rest... Those numbers can be majorly effected by one or two bad outings. 

I dunno, kinda seems like if anything you're improving the argument against using him on short rest. He's only had a few really bad outings ... but they've all come on short rest. Anyway, it seems like a pretty good idea in general to avoid running a 37-year-old out there on back-to-back nights, regardless of what the numbers say. 

[bizaff]

 

This is why I don't like mushing season stats together and coming up with averages for pitchers all that much. Sure, there is value there, but it misses a really important thing, imo.

 

Games are discrete events. I'll take the relief pitcher that gives up 10 runs in one game, and zero for the next 4, over the guy that gives up 2 runs in three games, and 1 run in the next two games (almost) any day of the year. I am NOT suggesting throwing out the bad games, but I think stats like what 538 did make more sense.

 

Agreed - I like the shutdown/meltdown statistics I've seen before.

 

Anyone watch his bad outings close enough to see if there was something specific?  It was over a spread of time so I'm guessing a single injury is unlikely..

Does anyone remember any specific problems he had in

[Dman]

 

I stumbled across this yesterday ...

 

Matt Belisle 2017

0 days rest: 8.22 ERA

1 day rest: 7.63 ERA

2+ days rest: 0.79 ERA

 

I think he'll continue to be very reliable as long as Molitor doesn't burn him out. 

 

Those stats jive with my memory.  I think he was over used early in the season.  I don't think that is the way Washington used him.  I think given proper usage he will perform better.  

 

Glad to see him turn things around.  Hope he keeps it up.

[Tom Froemming]

Also, Belisle had a fifth bad outing that was basically masked by an error. June 27 at Boston he gave up a three-run homer, but all the runs were unearned. Sure, it's tough luck that his defense let him down, but he still served one up.

 

Surprise, surprise that outing was also on zero days rest. So in 7.2 innings pitched on no rest, he's given up 10 runs (seven earned), has a 1.96 WHIP and opponents have hit .333/.405/.606 off him (1.011 OPS).

 

I'm not trying to dump on the guy or belittle his success, I'm just saying it's clearly a bad idea to have him pitch back-to-back days.

[Squirrel]

April 26 - 0.2 IP, 5 ER, 3 H, 2 BB (1 day rest)

May 7 - 0.1 IP, 6 ER, 4 H, 2 BB (1 day rest)

May 29 - 0.1 IP, 3 ER, 2 H, 1 BB (0 days rest)

June 11 - 0.2 IP, 4 ER, 4 H, 1 BB (0 days rest)

 

So again, those 0 days rest and 1 days rest prove to be pretty meaningless because they're so drastically affected by 2 outings. He has pitched in 8 games on 0 days rest, and 18 games on 1 day rest... Those numbers can be majorly effected by one or two bad outings.

I was going to ask what you think has changed or what Belisle might be doing differently that he seems to have 'righted the ship,' then saw what Tom wrote and thought, 'Okay that seems plausible,' then came to your post so now I'm asking ... is there something that's different or is he just finding his stride here?

[Riverbrian]

Relief pitchers are walking small sample sizes. 

 

They don't get enough innings to make up for anything horrific. 

 

 

[spinowner]

Cherry picking 4 games: interesting but of questionable validity

Separating based on rest: interesting and of higher validity

What would be even more interesting would be to see how the stats break down based on rest for MLB as a whole and for other specific pitchers.

[Squirrel]

Cherry picking 4 games: interesting but of questionable validity

Separating based on rest: interesting and of higher validity

What would be even more interesting would be to see how the stats break down based on rest for MLB as a whole and for other specific pitchers.

Knowing you, that would be a most excellent project for you!

[Vanimal46]

 

Seth if we're going to throw out the appearances where a reliever gets knocked around then literally every reliever in baseball is awesome.

 

Yep. It's a flawed way of thinking for sure... If I took away that +3 on holes 6, 10, 14, and 16, I'd be a professional golfer! 

[Seth Stohs]

 

Seth if we're going to throw out the appearances where a reliever gets knocked around then literally every reliever in baseball is awesome.

 

I think I noted that twice in the original post.

[Doomtints]

It would be interesting to look at the matchups he faced during those bad outings. There could be a clear indicator there of when not to use him other than a guess that he is tired.

[Dman]

 

Yep. It's a flawed way of thinking for sure... If I took away that +3 on holes 6, 10, 14, and 16, I'd be a professional golfer! 

 

Really all statistics are flawed.  Numbers can only tell us so much.  We can measure what happened and establish trends but we can never predict exactly what will happen in the future in baseball.

Personally I found the way Seth manipulated the stats helpful.  Belisle has gotten a bit of a bad rap after pitching poorly after being overused.  Taking out the few awful games he has had shows us he has been more consistent than most are giving him credit for.  While his averages don’t look great more often than not he has been successful in his role.

 

I think sometimes stats can take us down the wrong road.  Look at run differential.  Based on that you would think the Twins should be behind KC and Detroit and yet they are ahead of both teams.  So while run differential might be able to tell us the depth of talent a team has it can’t tell us what team might win a particular series very well at all.

 

While it is flawed to take Seth’s comparison to other pitchers without taking out their bad games,  I don’t think it is flawed at all in showing us just how consistently good Belisle has been in the majority of games he has played.

[AlwaysinModeration]

Seth if we're going to throw out the appearances where a reliever gets knocked around then literally every reliever in baseball is awesome.

Well, if we took every reliever and threw out their worst four outings, it would be interesting to compare. I think Belisle would come out looking pretty good.

 

Another way to do it: throw out the best and worst four outings. That would get us closer to his median, expected outing, I suppose. You could do that with all relievers and then compare them, and you might get something worth looking at.

 

I'm not going to do it, though.

[amjgt]

10% of his outing are complete crap.

90% of his outings are good to very good.

 

If people can't see that as valuable information in assessing a pitcher, then I don't know what to tell them

[DJSim22]

 

So, overall this year, Matt Belisle has pitched 45.2 innings in 47 games. He's got a 4.34 ERA and a 1.27 WHIP.

 

Now, he's had four HORRIBLE games:

 

1. April 26 - 0.2 IP, 5 ER, 3 H, 2 BB
2. May 7 - 0.1 IP, 6 ER, 4 H, 2 BB
3. May 29 - 0.1 IP, 3 ER, 2 H, 1 BB
4. June 11 - 0.2 IP, 4 ER, 4 H, 1 BB

 

So, we all know that you can't just pick and choose which games to count in numbers, but just for fun, I thought I'd do it anyway.

 

So, in the 43 games that Belisle has pitched in (not counting the 4 above), he has thrown 43.2 innings. He has a 0.82 ERA and a 0.89 WHIP! That's pretty crazy!

 

 

Those four games... 2 IP, 81.00 ERA, 8.50 WHIP.  Also, pretty crazy!

 

He hasn't given up an earned run since June 26 (16 outings), and a run period since June 27 (14 outings). An inherited runner hasn't scored off of him since July 1st. Only 4 of 23  (17%) of inherited runners have scored on him all season.

 

In other words, he's been great this year... except for those four games... (and just to repeat, I'm not saying we can just eliminate the bad stuff, but perspective is good too)

 

I've been trying to argue this for over a month.  Alas, you did a much better job of it.  If anything it show consistency.

 

[snepp]

 

Relief pitchers are walking small sample sizes. 

 

They don't get enough innings to make up for anything horrific. 

 

John Smoltz, 2002, gives up 8 earned runs in his second relief appearance of the year in 2/3rd of an inning. It takes him until the end of July to lower his ERA below 4 despite giving up a mid-.500 OPS during that time.

[DocBauer]

Not going to jump in to the numbers directly, but what I find interesting is that the HORRIBLE numbers/performances that Seth brought up occurred in April, May x 2, and early June. All in the first half of the season, and 3 in the first 2 months of the season.

[Darius]

I'd imagine you could find similar trends with any pitcher with bad numbers (90% of their innings are great, skewed by a handful of bad ones).