Article: Trying to Get a Reed on Addison

[Jamie Cameron]

Twins fans were thrilled when they signed Addison Reed this offseason on a two year deal worth approximately $16 million. Given the nature of the free agent relief pitching market in recent years, the Reed signing seemed not only to be a great get for the Twins, but a significant bargain given the deals that Rockies handed out to Bryan Shaw, Greg Holland and Jake McGee. The Twins, it seemed, had obtained a reliable reliever with a proven track record who could take on a high workload of high leverage innings and support Fernando Rodney at the back end of the bullpen.Wednesday night in Detroit was the culmination of a rough few weeks for Reed. Over his last 15 appearances (13.2 innings), he’s held a 5.27 ERA, 1.68 WHIP, and surrendered some key leads. Despite his season-long numbers being relatively in line with his career figures, any kind of digging into Reed’s numbers this season lead to an inevitable conclusion: he’s in rapid decline.

 

Peripheral Stats

Looking at the last 3 years lead to some worrying conclusions about Reed’s early season form with the Twins.

Download attachment: ReedGraphic.png

It’s worth pointing out that Reed and Ryan Pressly have been significantly over-worked by Paul Molitor this season in a similar fashion to Trevor Hildenberger in 2017. Both have taken significant steps backwards in the past few weeks. While Pressly’s peripheral numbers line up almost exactly with his current performance, the same cannot be said for Reed. While overuse is certainly a factor to consider with Reed, his effectiveness is also quite clearly decreasing. Reed has never been a ground ball pitcher, but everything he throws this year is getting hit in the air, with an increasing fly ball and HR rate, it has led to him being less effective, and more noticeably, in high leverage situations.

 

Let’s talk about WPA for a second. No it’s not a bougie craft beer, win probability added measures how much a given player increases or decreases their team's probability of winning. Reed has been especially brutal in high leverage situations. His figure of -0.62 is already two thirds of a win below league average. He has the 21st worst WPA in the entire league, right behind Blaine Boyer of the Royals (11.60 ERA). That is a problem for a pitcher Molitor frequently relies on as his ‘eighth inning guy’ who is constantly pitching in high leverage situations with the game on the line.

 

Velocity

 

Reed’s velocity has also been in steady decline over the last few years. In the last 3 years, Reed’s average fastball velocity has fallen from 93.2 mph (2016), to 92.8 mph (2017), to 91.8 mph (2018). While velocity isn’t everything, consider this; Reed’s fastball was worth 12.9 runs above average in 2017, through roughly one third of the season in 2018, it’s worth 0.5 runs above average. Reed’s decline in velocity, combined with a decreasing number of ground balls and no Byron Buxton, is hurting Reed’s 2018 effectiveness significantly.

 

Moving Forwards

 

It’s fair to say I’m picking on Reed in the middle of a rough patch. It’s also true that a stagnant Twins offense could have reduced his disappointing WPA by scoring more freely (or at all) against sub-standard opposing pitchers. The Twins bullpen in general has been poor in high leverage situations (not aided by very little margin for error). Moving forward I’d argue that Trevor Hildenberger should be pitching in as many close and late situations as possible. He the Twins’ leader in WPA at 0.37 (68th best in the league). Since struggling in the first month of the season, Hildy has been rolling. In May and June, he’s given up a .347 SLG and 3 ER in 20.2 IP. Reed can undoubtedly be effective contributor for the Twins over the next season and a half, but Minnesota would do well to move him to a less impactful spot in a bullpen which this year has very little margin for error.

 

Click here to view the article

[Dave The Dastardly]

Clever headline and nice analysis.

[nicksaviking]

I think that diminishing velocity likely had a lot to do with why Reed looked like a steal despite his past results. If he was throwing 95-97 even with slightly inferior stats, he likely would have had other teams lining up to pay him more.

 

The Twins rightfully were applauded for making a ton of offseason moves, but nearly all were considered bargains. Now with all the data readily available to all teams it's not like the Twins can see something that other teams don't. I think the Twins are going to have to realize that waiting for bargains isn't the way to go, they're going to have to start overpaying for the guys who are truly difference makers.

[Platoon]

It's very easy to say that Reed (and Pressly) are suffering from Molitor Abuse. As will Hildy if Molitor switches his focus to him. Then again it's easy to say that Reed is simply losing it. But it's more likely a combination of the two. An astute MOY would realize that one good Reed outing is better than two bad ones. This pattern of overuse is not recent, driven by the need to remain in 'the race'. He burned out relievers in 2015 also, in the race for the first draft choice. I think that the high pitch counts for SP, and the 100 pitch plateau is causing a lot of stress on bullpens, making decisions on how to distribute innings more complicated, and more impactful. And the implications of the decisions even more obvious.

[Thrylos]

Velo is one thing, movement is another.  If you look at the pitchf/x movement data, his fastball has flatten out this season, which makes it less effective, and I suspect is partially the reason for the decrease in GB.

 

As far as overuse goes, the previous 2 seasons he appeared in 49 and 48% of his teams' games respectively.  So far this season he has appeared in 49% of the Twins' games.  Not that different usage in the past.

 

Also, if you look at the past in 2016 he had an xFIP of 2.89, SIERA of 2.56, in 2017 3.85 FIP and 3.35 SIERA, which both are about where his ERA is at 2018.  xFIP and SIERA being leading indicators (and sitting at 4.83 and 3.97 respectively right now) does not bode well for Reed.  Maybe that reflects the size of the contract his signed, because he is more like a 6th or 7th inning pitcher.

 

The problem however is, the still he likely is the best pitcher in this pen, and that is no small reason why the Twins' pen is 11th in the AL in ERA, 13th in FIP, and 12th in WPA,  tied for second for the most BS and L, just a game behind the leader in that category...

 

The pen has an 11-16 record, the rotation 18-20.  And Reed has an 1-5 record.  And, unlike for starters, L for relievers is a more meaningful metric.

[jimbo92107]

It's the age-old story. Velocity x Command = Effectiveness. As V goes down, C better go up, or E will go down rapidly. Unfortunately, over Time, both V and C tend to go down, unless you add a new pitch, Pn. This gives

 

((VC)Pn)/T = E

 

Clearly Reed's V and C have been suffering, along with over-use, U+. Thus

 

((VC)Pn)/(T+U) = E

 

Age-old story.

[RatherBeGolfing]

 

 

 

The problem however is, the still he likely is the best pitcher in this pen

 

This right here. This is why this is a problem. We just don't have enough bullpen depth where we can sit someone like Reed for a while and throw others out there to eat up some innings.

[AceWrigley]

I think we do have enough bullpen depth, its just that some of it is in Rochester.

[Shaitan]

Eh, pitchers go through hot/cold streaks. Not worried. (Though Molitor doesn't have to have him pitch 5 of every 7 days in tight situations as he works through it.)

[Tomj14]

 

I think we do have enough bullpen depth, its just that some of it is in Rochester.

I am not sure it is a depth issue.

Molitor needs to realize that the starters as a team are averaging less than 5 2/3, and he needs a plan every night for the bullpen to pitch 3 to 5 innings. IMO he doesn't go into the game with that plan and reacts to the game situation and puts Pressley and Reed in almost every game. Then Duke and Hildy fill the gaps.

Why have not have a plan for Magil to pitch the 5th and/or 6th in games Odorizzi and Romero start or something other than the starter gets in trouble in the 5th bring in Pressly to finish that inning start the 6th, matchup up Duke and Hildy for the 7th, Reed 8th every game and hope Rodney saves the game.

[Franz]

During Reed's last outing, the 4-run eighth inning against Detroit, his fastball was clocking in at 89-90 mph while his slider was at 84-85. Not much difference in velocity between those pitches. Perhaps if Reed is losing fastball velocity, he needs to work on more separation between those pitch speeds to keep hitters off balance?

 

Does he have anything besides fastball/slider??

[Fritzderkat]

After watching him the past couple of weeks I've come up with a nickname: Additive Reed: He makes their offense go.

[Jacksson]

In Reed's most recent failure, his FB was at 90.

In Pressly's last bad outing he blew away the first batter - striking him out on 3 pitches that featured 97 MPH Heat. For whatever reason, he then went to curves and sliders which the opposition teed off on. Whether it was Pressly calling off the catcher's signs or the catcher calling a bad game: when you are throwing at 97 you keep feeding them that until they show that they can hit it. Don't back away from your best pitch trying to be cute.

[h2oface]

Trevor Hildenberger just recently gave up another 2/2 inherited runners scored to add to the 7/7 stretch at the beginning of the year....... which is not thst far behind us.... to make 9 that I can remember without looking them up to see if there is more. 

 

Inherited runners scored, to me, is the most important stat to indicate the effectiveness of a reliever. It is, after all, supposed to be relief, and not let the runs be charged to someone else, and I still look OK with the other stats. I certainly don't think Hildenberger is the long term answer, but he is what we have.

[stuck0]

Correction: Rockies signed Shaw, McGee, and Wade Davis this off season (not Holland).

[ashbury]

Inherited runners scored, to me, is the most important stat to indicate the effectiveness of a reliever. It is, after all, supposed to be relief, and not let the runs be charged to someone else, and I still look OK with the other stats. I certainly don't think Hildenberger is the long term answer, but he is what we have.

MLB average so far this year is 32% of inherited runners score (last year 30%). The Twins team-wide are at 42% (last year 28%) - down at the bottom of the majors but tied with a few others. Hildenberger at 65% and Pressly at 42% are the main culprits. Reed's at 25% for the season.

 

It's kind of a crude stat, combining runners' locations and numbers of outs, but still indicates where some of our pain is coming from. A league-average rate would account for about 10 fewer runs, or pro-rated to a full season around 25 if we continued like this. By a common rule of thumb, that would be 2.5 losses - but given all the close games we've had, the leverage is for more than that. Maybe it's just Small Sample Size at this point in the season, but the trend needs to stop.

[mikelink45]

I think that there is an issue with managing the new system - relief pitchers were not regularly schedule from inning 5 on and while a few bullpens have accumulated the arms to handle that demand, there are not enough to go around.  I do not envy Molitor.  If we need at least four RP per game, Taylor, Belisle, Magill, Duke, Pressly, Hildenberger, Rodney, Reed, and Pressly.  We do not want to overuse them so which ones do we use.  Put four together and then another four and then another four.   No one is going to pitch 9 so we have to keep calling on them.  Yes we have good arms in the minors, but to do it right we have to keep calling them up and sending them down.  This era has created a problem, but no answer.  I do not blame Molly for this. 

[Jham]

I think that there is an issue with managing the new system - relief pitchers were not regularly schedule from inning 5 on and while a few bullpens have accumulated the arms to handle that demand, there are not enough to go around. I do not envy Molitor. If we need at least four RP per game, Taylor, Belisle, Magill, Duke, Pressly, Hildenberger, Rodney, Reed, and Pressly. We do not want to overuse them so which ones do we use. Put four together and then another four and then another four. No one is going to pitch 9 so we have to keep calling on them. Yes we have good arms in the minors, but to do it right we have to keep calling them up and sending them down. This era has created a problem, but no answer. I do not blame Molly for this.

But you just suggested a solution: take advantage of your player options to extend the number of available arms in your pen. We have a number of available major league capable arms who can be shuttled in and out and kept fresh. Problem is when they're called up, Molly doesn't use them anyway. More often them not, it seems they're sent back down to get some work rather than switch out for a fresh arm.

[The_Phantom]

This isn’t a complicated issue at all. Molitor strictly adheres to the magic 100 pitch count notion, even if a starter is at something like 5.2 IP and only allowed one run. He very, very rarely stretches the starters beyond that. Whatever batter faces the 100th pitch, that’s the last batter the starter sees. So our bullpen, especially Reed, is horrendously overtaxed.

[Doctor Wu]

 

The Twins rightfully were applauded for making a ton of offseason moves, but nearly all were considered bargains. Now with all the data readily available to all teams it's not like the Twins can see something that other teams don't. I think the Twins are going to have to realize that waiting for bargains isn't the way to go, they're going to have to start overpaying for the guys who are truly difference makers.

I have to agree with that. Most of the time those "bargains" were that way for a reason ... or three. If they Twins want to compete with the elite teams they will have to either draft better/wiser, or dish out more money for the top tier free agents.

[David HK]

 

It's very easy to say that Reed (and Pressly) are suffering from Molitor Abuse. As will Hildy if Molitor switches his focus to him. Then again it's easy to say that Reed is simply losing it. But it's more likely a combination of the two. An astute MOY would realize that one good Reed outing is better than two bad ones. This pattern of overuse is not recent, driven by the need to remain in 'the race'. He burned out relievers in 2015 also, in the race for the first draft choice. I think that the high pitch counts for SP, and the 100 pitch plateau is causing a lot of stress on bullpens, making decisions on how to distribute innings more complicated, and more impactful. And the implications of the decisions even more obvious.

I concur. 

 

While ARE's usage stats are in line with previous seasons, he is a year older every year-- the body simply needs more recovery time as it ages. Couple that with the fact that not only are ARE and RPE being overused in the same spots again and again, blowing leads again and again, there seems to be no recognition on the part of the MoY that maybe we should try something different this game, or the next.

 

Same script, same results, same string of losses to whom we previously considered inferior teams. Especially so when the offense is as offensive as it has been, those late innings are unfortunately crucial nearly every time out these days.

 

So what if we have to bounce Duff, Curtiss, Busenitz, et.al. back and forth a few times? It gives them a taste, too, and gives those other guys the break they need here and there. Plenty of other teams do it. Get the hot hand in Roch up in the Cities. If he fries some late innings, well, bump him back down, and he knows what he needs to work on going forward. At least if we're giving away games, let's do it with some potential benefit.

 

I'm just speculating, but Ozzie Guillen would probably do it that way.

[Danchat]

 

This isn’t a complicated issue at all. Molitor strictly adheres to the magic 100 pitch count notion, even if a starter is at something like 5.2 IP and only allowed one run. He very, very rarely stretches the starters beyond that. Whatever batter faces the 100th pitch, that’s the last batter the starter sees. So our bullpen, especially Reed, is horrendously overtaxed.

Well, the main reason Molitor pulls guys around 100 pitches is because that's when they typically wear out and start struggling. How many times have we seen Molitor send a starter out to start the 6th or 7th and see them allow baserunners just to pull them? Berrios is the only one I'd trust to be effective past 100 pitches.

[Riverbrian]

Early in the year I was watching Hildenberger get rocked and I was saying that we should send him down to AAA for awhile to get him back to where he was. Now I'm thinking that Hildy is looking pretty good. 

 

Early in the year... I was watching Reed hit the catcher's target dead center on every pitch and he was looking real impressive to me and I was thinking we got a nice bargain. Now I'm thinking... OMG... Did we pay too much for Reed?

 

I think it comes down to: That's the way it goes in the bullpen.

 

Bullpen guys are small sample size on top of small sample size adding up to a small sample size. The guys in the bullpen have to be nearly perfect to avoid the criticism. 

 

If a starter gives up a run in the first... he gets 4 or 5 more innings to hang some zeros to mitigate that run in the first. If a bullpen guy gives up a run he typically doesn't get another inning that game and that run in one inning is a 9.00 in the books and it's gonna hang there for a couple of days before he gets another inning sometimes. 

 

Overuse? I don't know... I guess that depends on what each individual can handle and I don't know the answer to that. 

 

Hildenberger 34.2 IP

Pressly 34.1 IP

Reed 33.1 IP

 

And when you compare his usage to his bullpen peers from other teams. There is quite a lengthy list of bullpen arms in the 30 to 35 IP range. 

 

Over the past 30 days:

Hildenberger 15.0

Pressly 11.0

Reed 11.0

 

And once again... there is a lengthy list of bullpen peers from other teams in the 11 to 15 IP range across the league in the past 30 days. 

 

I think it's possible that the overuse claim is an assumption based on recent performance but not necessarily correct. 

 

During the off-season... Reed is one of the guys that I wanted the team to sign. I'm hoping he'll pull of a string of good small sample size performances to balance his most recent small sample size performances. 

 

 

[yarnivek1972]

Pressly is still on pace to appear in almost 90 games. No one in MLB appeared in even 80 last year. Last year the highest game total was Rogers’ 69 and the most bullpen innings were Duffey’s 71. The Twins have THREE guys on pace to pitch over 70 bullpen innings. Molitor actually eased up last week. Of course, part of that was some non-competitive games. Going into last week, the situation was even more ridiculous.

[Twodogs]

Part of the problem could be that the Twins have been in a ton of close late games this year. It hasn't helped that they have come on the wrong side of many of those games. The Twins need to bust out and blow a few teams out of the water to get guys like Magil in there more often. If they had won more of those close 1 run games they'd be in a much better spot. So hopefully they can have a few more of those 9 - 3 games like the other day so as to rest those bullpen guys like Pressly, Hildy, Reed etc...

[yarnivek1972]

Part of the problem could be that the Twins have been in a ton of close late games this year. It hasn't helped that they have come on the wrong side of many of those games. The Twins need to bust out and blow a few teams out of the water to get guys like Magil in there more often. If they had won more of those close 1 run games they'd be in a much better spot. So hopefully they can have a few more of those 9 - 3 games like the other day so as to rest those bullpen guys like Pressly, Hildy, Reed etc...

Or maybe use Magill in those closer games until he proves he isn’t capable of doing so?

 

The guys Molitor is using now are already losing games. What difference does it make if it is Reed losing or Magill?

[Mike Sixel]

 

But you just suggested a solution: take advantage of your player options to extend the number of available arms in your pen. We have a number of available major league capable arms who can be shuttled in and out and kept fresh. Problem is when they're called up, Molly doesn't use them anyway. More often them not, it seems they're sent back down to get some work rather than switch out for a fresh arm.

 

that's on the FO, and their filling the RP list with veterans w/o options.....something many people here suggested was a problem with their plan, and that yes, one year deals for RPs can be bad....

[jorgenswest]

I would join you with the bullpen being on management. However, I don’t think the data supports an option problem. Duffey, Slegers, Busenitz, Curtis, Moya, Littell and Magill have been optioned one way or the other. Additionally I think Rogers and Hildenberger still have options.

 

I think the problem is lack of talent relative to contending teams.

 

That might stem from...

 

- Inability to recognize and utilize talent of minor leaguers

- inability to develop that talent

- poor assessment of the talents of veteran free agents

- poor use of the talent at the major league level

 

Certainly any expectation that Kinley and Hughes would contribute to an effective bullpen of a contending team was unreasonable.

[Twodogs]

Or maybe use Magill in those closer games until he proves he isn’t capable of doing so?

 

The guys Molitor is using now are already losing games. What difference does it make if it is Reed losing or Magill?

Yeah, I think that they need to use all 7 - 8 arms in the pen, no sense in using the same 3 or 4 all the time. On pace for 90 games is a lot.

[Brock Beauchamp]

 

Trevor Hildenberger just recently gave up another 2/2 inherited runners scored to add to the 7/7 stretch at the beginning of the year....... which is not thst far behind us.... to make 9 that I can remember without looking them up to see if there is more. 

 

Inherited runners scored, to me, is the most important stat to indicate the effectiveness of a reliever. It is, after all, supposed to be relief, and not let the runs be charged to someone else, and I still look OK with the other stats. I certainly don't think Hildenberger is the long term answer, but he is what we have.

Inherited runners are really important but as a stat, it's pretty flawed. If you inherit a runner on first with 1-2 outs and let them score, that's bad pitching.

 

If you inherit runners on second and third with no outs and both score, that's not really on the reliever (maybe you can make a case that the runner on second should only score ~50% of the time but the runner on third is a gimme).

 

It's not terribly different than judging a starter on his win total. Sure, the outliers are pretty solid indications of good/bad (if a starter wins 20, he's very good... if he loses 15+, he's pretty bad) but the fat middle part of the bell curve is pretty hard to judge on that stat alone.