Mediocre Starter vs Late Inning Reliever Discussion Thread
by Brad on Jan 31, 2012 2:30 PM CST
Email
PrintAfter acquiring Yu Darvish, the Rangers have seven legitimate candidates for the starting rotation in Colby Lewis, Derek Holland, Matt Harrison, Neftali Feliz, Alexei Ogando and Scott Feldman. They also added two top starting pitching prospects, Martin Perez and Neil Ramirez, to their 40 man roster this winter making them potential candidates for the 8th and 9th spots on the depth chart. And if all that wasn't enough (I imagine the limes guy with JD's face and the caption, "Why can't I hold all these starting pitchers?") the Rangers have been linked to Roy Oswalt as a potential candidate to join this already crowded group.
The discussions around potentially moving one of the young arms back to the bullpen made me realize that I think I would rather see a pitcher that I thought might have a ceiling as a mediocre starter become a late inning reliever if I thought he could be elite in that role. This prompted me to take a look at the last 10 years and start classifying starters and relievers to see how many pitchers fall in each bucket and what some of their attributes look like.
I counted any season in the Starter category if they had at least 20 games started or 90 IP as a starter. Relievers had to have had at least 40 IP as a reliever in a season. From there I used their fWAR for the season to put them into one of five tiers.
| Starters | Num | Pct | Age | IP | WAR | ERA | FIP | K% | BB% | HR% | SwStr% |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 1. 5.5 or greater | 88 | 6.33% | 28.9 | 225.8 | 6.7 | 2.92 | 2.99 | 22.9% | 5.8% | 1.9% | 10.6% |
| 2. 4.0 to 5.5 | 164 | 11.79% | 28.3 | 207.9 | 4.7 | 3.55 | 3.57 | 19.4% | 6.9% | 2.1% | 9.5% |
| 3. 2.5 to 4.0 | 353 | 25.38% | 28.1 | 187.7 | 3.2 | 3.95 | 3.98 | 17.6% | 7.3% | 2.5% | 8.8% |
| 4. 1.0 to 2.5 | 488 | 35.08% | 28.3 | 161.2 | 1.8 | 4.51 | 4.50 | 15.4% | 8.1% | 2.8% | 7.9% |
| 5. Less than 1.0 | 298 | 21.42% | 28.3 | 139.1 | 0.4 | 5.16 | 5.18 | 13.6% | 8.8% | 3.5% | 7.4% |
| Totals | 1391 | 100.00% | 28.3 | 172.8 | 2.5 | 4.20 | 4.22 | 16.8% | 7.6% | 2.7% | 8.5% |
| Relivers | Num | Pct | Age | IP | WAR | ERA | FIP | K% | BB% | HR% | SwStr% |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 1. 2.0 or greater | 94 | 5.40% | 29.5 | 77.6 | 2.6 | 2.29 | 2.43 | 28.9% | 7.3% | 1.3% | 13.1% |
| 2. 1.5 to 2.0 | 101 | 5.80% | 29.3 | 74.4 | 1.8 | 2.79 | 3.01 | 24.3% | 7.8% | 1.7% | 11.9% |
| 3. 1.0 to 1.5 | 226 | 12.99% | 29.8 | 70.0 | 1.3 | 3.08 | 3.25 | 21.9% | 8.0% | 1.7% | 10.7% |
| 4. 0.5 to 1.0 | 381 | 21.90% | 29.6 | 65.6 | 0.8 | 3.48 | 3.72 | 20.0% | 9.0% | 1.9% | 10.0% |
| 5. Less than 0.5 | 938 | 53.91% | 29.8 | 59.6 | 0.0 | 4.35 | 4.63 | 17.2% | 9.9% | 2.9% | 9.3% |
| Totals | 1740 | 100.00% | 29.7 | 64.1 | 0.6 | 3.73 | 3.98 | 19.6% | 9.1% | 2.3% | 10.1% |
Using these standards, the past 10 years has produced 421 seasons of a reliever with at least 1.0 WAR and 605 seasons of a starter with at least 2.5 WAR. If you go by WAR alone to determine value, a 2.5 WAR starter would seem more valuable than a 1.0 WAR reliever, but a 1.0 WAR reliever is more scarce and that scarcity is magnified as you increase to 1.5 or 2.0 WAR seasons for a reliever. Part of the challenge in evaluating the value of a reliever is properly accounting for the higher leverage of late inning situations. Even though relievers pitch less innings, their impact to the actual win loss record of a club may be more meaningful than WAR currently indicates (a known discussion point among advanced metric enthusiasts).
Another element of information is the pitch profile of the different tiers of starters and relievers in terms of velocity and pitch selection.
| Starters | FBv | FB% | CT% | SL% | CB% | CH% | KN% | SF% |
|---|---|---|---|---|---|---|---|---|
| 1. 5.5 or greater | 92.06 | 58.1% | 5.4% | 11.9% | 12.1% | 10.3% | 0.0% | 2.2% |
| 2. 4.0 to 5.5 | 91.23 | 60.8% | 3.3% | 12.1% | 11.0% | 10.7% | 0.0% | 2.1% |
| 3. 2.5 to 4.0 | 90.43 | 59.8% | 3.1% | 12.4% | 10.8% | 11.5% | 1.2% | 1.3% |
| 4. 1.0 to 2.5 | 89.53 | 59.3% | 3.0% | 12.7% | 10.0% | 12.5% | 1.2% | 1.3% |
| 5. Less than 1.0 | 89.26 | 60.2% | 2.9% | 12.3% | 10.7% | 12.4% | 0.5% | 1.0% |
| Totals | 90.16 | 59.7% | 3.2% | 12.4% | 10.6% | 11.8% | 0.8% | 1.4% |
| Relivers | FBv | FB% | CT% | SL% | CB% | CH% | KN% | SF% |
|---|---|---|---|---|---|---|---|---|
| 1. 2.0 or greater | 93.42 | 64.2% | 6.0% | 19.5% | 4.6% | 4.3% | 0.0% | 1.3% |
| 2. 1.5 to 2.0 | 92.46 | 65.2% | 2.5% | 17.9% | 5.1% | 7.9% | 0.0% | 1.4% |
| 3. 1.0 to 1.5 | 91.56 | 64.3% | 1.8% | 17.9% | 6.4% | 7.9% | 0.0% | 1.7% |
| 4. 0.5 to 1.0 | 91.65 | 64.1% | 1.8% | 18.2% | 6.4% | 7.4% | 0.0% | 2.2% |
| 5. Less than 0.5 | 91.01 | 62.9% | 2.3% | 17.9% | 6.1% | 9.0% | 0.2% | 1.7% |
| Totals | 91.48 | 63.6% | 2.4% | 18.1% | 6.0% | 8.1% | 0.1% | 1.8% |
These numbers show the increased use of fastball/slider combination among relievers compared to starters and their use of more diverse secondary pitches. Though I think it's important to note, the difference between a potential starter and reliever is more than just their ability to throw certain pitches. Each pitchers mechanics and body type has an impact on their durability and ability to manage a starter's workload as well.
For the discussion:
1) Assuming the need for starters and late inning relievers is equal for your team's pitching staff and which capacity would you rather use a pitcher with a 2.5 to 3.0 WAR ceiling as a starter and a 1.5 to 2.0 ceiling as a reliever?
2) When trying to determine the value of a reliever versus a starter, what elements do you consider and give priority?
3) What is your ideal usage of the Rangers batch of young potential starters: Holland, Harrison, Feliz and Ogando?
28 comments
|
0 recs |
Do you like this story?
Comments
my 2 cents
1) I’d prioritize having 5 quality starters first. A decent closer can always be found, and a QSP, not so easy.
2) A starter that can regularly go 7 innings versus 6 is a big defining point between valuable vs. very valuable. For a closer, it’s just all about getting it done. I suppose the ability to go back-to-back days might be a dividing line.
3) Holland, Harrison, are starters, no ifs ands or buts. Feliz should get his chance. I think he’ll succeed, I’m just not sure he’s ready opening day. Ogando I can see being at a minimum, capable of being a shut down reliever in any inning, or a good SP with the potential of being am elite SP if he can master his CU.
Baseball's hard, guys. I mean, it really is. You can love it but, believe me, it don't always love you back. It's kind of like dating a German chick, you know?
The save is the most important stat ever
/sarcasm font
I hate The Angels so much, it's actually cramping my style. (See RevHaloFans SBN profile page)
by Gay For Feliz on Jan 31, 2012 2:51 PM CST via mobile reply actions
.
If you go by WAR alone to determine value, a 2.5 WAR starter would seem more valuable than a 1.0 WAR reliever, but a 1.0 WAR reliever is more scarce and that scarcity is magnified as you increase to 1.5 or 2.0 WAR seasons for a reliever.
I’m not sure this comparison provides information on (potential) scarcity of types of pitchers, as opposed to the relative frequency of certain kinds of seasons by certain types of pitchers.
Garoon: Banned on LSB. Available Only Via Supplement.
Good point.
There are certainly elite starting pitchers who if put in the reliever role would be able to be elite relievers. I had that thought, but didn’t make the point.
"I became an optimist when I discovered that I wasn't going to win any more games by being anything else." by Earl Weaver
Do you make these tables in SBNation's tool or do you import them from somewhere else?
Ever been in a boardroom, or a gang fight? Ever save a life? Ever won a court case against the odds? Ever held a dying buddy in your arms?.- jackanape on "perspective," 11/8/2010.
"do you a jet ski rental place would work here" - Mike E, 5/10/2011.
I use Excel to make them.
I use CONCATENATE formulas to build the HTML and the TEXT function in Excel to make numbers formatted the way I want.
"I became an optimist when I discovered that I wasn't going to win any more games by being anything else." by Earl Weaver
Well shit.
I don’t know what that means.
Ever been in a boardroom, or a gang fight? Ever save a life? Ever won a court case against the odds? Ever held a dying buddy in your arms?.- jackanape on "perspective," 11/8/2010.
"do you a jet ski rental place would work here" - Mike E, 5/10/2011.
He knows the HTML code underlying the table
and so all he has to do is switch out the numbers in the table in the HTML code. He uses Excel to do the plug and play of it, essentially.
You should post the excel file, match.
I typically make a new one every time.
But I’ll make a sample version and post it.
"I became an optimist when I discovered that I wasn't going to win any more games by being anything else." by Earl Weaver
I should really do things this way, instead of the hard way.
Garoon: Banned on LSB. Available Only Via Supplement.
I started it when I was doing the Community Projection reviews.
I did the first two by hand and then got tired of that shit.
"I became an optimist when I discovered that I wasn't going to win any more games by being anything else." by Earl Weaver
I just realized you wrote this and not Adam.
I’m going to link to it from the regional site in a bit.
SB Nation Dallas-Ft. Worth - Christopher Fittz is better than porn!
I'd view Ogando as having the ceiling of a Tier 2 SP and Tier 1 or 2 RP
My problem with using WAR in this scenario for RP’s is that a RP’s fWAR is very dependent on the manager’s use of him in high-leverage situations.
We’ve personally seen Wash use talented RP’s in low-leverage situations, greatly limiting their WAR.
Just a hunch, and maybe you could easily calculate this, match
But, I’d think there’d be much less variance b/w a SP’s FIP and fWAR than a RP’s FIP and fWAR.
Ogando, for instance, was tier-2 by FIP, but tier-4 by fWAR
IP comes into play.
And the number of innings you think a particular pitcher can pitch in a full season is relevant.
"I became an optimist when I discovered that I wasn't going to win any more games by being anything else." by Earl Weaver
I'd have to see what their PEACE was first.
"Look if you want a journalist hire a UT grad. If you want a doctor you pick a Baylor grad. If you need a vet you see an Aggie grad. And if you need a pizza you call a Tech grad." -BUBacker
Beware of Trolls: b.pate, jonthefon, K.S.B., Panorama, Robert L. Bishop, TideNTexas
by Mike E on Jan 31, 2012 3:21 PM CST via Android app reply actions
Elite
Seems to me that you’d want elite players at every position (a pipe dream, I know). So, if it came down to being elite at one spot and mediocre at another, I’d go for elite.
So you openly admit that WAR is not a great way to measure reliever value yet thats how we are supposed to compare a great reliever to a mediocre starter?
Not criticizing just curious.
Great stuff as always breaking things down.
As for your questions
1: Depends on a few things. what role would the reliever be in? Closers, as much hate as the name gets, are very important pieces and have the ability to really affect a teams W/L record depending on their performance. If your reliever comes in the 6th inning and gives up the lead you still have 3-4 at bats to make up that deficit whereas if your “closer” coughs up a lead you only have 1 at most.
My other question being how do they accumulate that WAR as a starter? Is it a guy who goes out and puts up great FIP type numbers (K’s, low BB’s, etc) yet cant go deep into games or is it a guy who doesn’t put up great FIP numbers yet gives you innings?
2: I guess 1 answers 2 for me
3: Well since the switch has already been made with Feliz i’ll stick with that. I have long been a guy who has little faith in Feliz’s ability as a starter to be the TORP a lot on here believe he will be. But given that the organization has already decided to go with him for 2012 at least I would go with him and the two lefties. Ogando goes to the pen as a late game fireman type guy who can come in anywhere from the 5th to the 8th and give you anywhere from 1-3 innings (depending on prior usage obviously). After 2012 if Feliz stays in the rotation then Ogando steps into the role Adams will play for us in 2012 and possibly takes over for Nathan.
I think WAR is lacking in showing the actual value of a reliever.
We make fun of teams for paying for the “save” statistic… but even if you want to just use FIP or other metrics to indicate who the elite relievers are… the current $/WAR in FA seems skewed for how teams value late inning closers.
And I’m not saying that you should use WAR to compare a mediocre starter to a great reliever… I did make the comparison and then showed some things that make me think you shouldn’t.
"I became an optimist when I discovered that I wasn't going to win any more games by being anything else." by Earl Weaver
there just isn't a good way to value RPs
either to each other and certainly not to SPs
by RangerMad on Jan 31, 2012 6:44 PM CST via Android app up reply actions
Good work
However, the WAR ranges that you have used to define pitcher types appears too restrictive. You seem to be using an unstated definition of talent (and the resulting scarcity generated by your first assumption) here I guess. Do you think it is reasonable (or useful, or whatever) to define the universe of TORPS as a mere 25 pitchers in any given season? I don’t.
If you just take quick glance at the last 2 years, the top 30 WAR figures begin at around 4, while ACE production might be said to begin around 5.5. Top 60 WAR figures begin about 2.6. I"m in favor of defining talent much more in line with actual production. The only outliers will be Aces and below average starters, but that only really affects the top 10 and bottom 15 if the universe of pitchers starts at 150. Over the last 2 years, the average # of pitchers who make about 15GS/100ip with an sub-5 FIP is about 140 which seems to be the (average) “actually existing universe” of pitching talent which should govern the definition of any ideal (SP) type, if the type itself is to have any heuristic value at all.
It seems to me the actually existing universe of pitchers over the last few years or so based on production might be said to look something like this (or at least this is what I use):
ACE: ~5.5+
#1: ~4-5.5; FIP: 2.4-3.5
#2: ~2.6-4; 3.5-3.85
#3: ~2-2.6; 3.88-4.2
#4: ~1-2; 4.2-4.6
#5: ~.5-1 (top 20 FIP sits 4.6-5)
#6: ~.4 or less (bottom 10 sits 5-5.7)
So theres all that. However, if you want to use a much more restrictive definition of ideal types for SP based on some definitions pivoting on a talent-scarcity scale (or desire – what makes scouts drool), I think that is a much more complicated (and interesting) discussion. It would mean that you would have to justify the definition of 83% of the leagues pitchers as unTORP, that is to say, undervaluing everybody else. I guess you could do that but I’m not sure why you need to.
The central challenge here, as I see it, is a counterfactual:
If a given starting pitcher/group of starting pitchers with a stated WAR range was, in a different universe, a relief pitcher/group of relief pitchers, what WAR (range) would that reliever/group of relievers post?
Put another way: what sort of relief performance would be predicted from a given “performance type” of starting pitcher?
Garoon: Banned on LSB. Available Only Via Supplement.
Yeah
I guess I was kind of ignoring the problematic in general while inveigling an altogether different discussion.
Ok
At the risk of drawing your ire, this is like a running gag, right? Do you really think that sentence is the most descriptive and efficient way to convey the point? I honestly feel a little dumb for being 75% sure I’m missing the joke.
by ab03 on Feb 1, 2012 8:51 AM CST via mobile up reply actions
Huh?
none of the above. I was acknowledging the fact that I skimmed over the central problematic in Stix’s (if-then) post, as Snark pointed out very succinctly, in favor of focusing on another issue which Stix brought up about a week ago in passing (defining starter types by WAR) which had been on my mind. I define starter types more loosely as I indicated above, which is different that Stix. Nothing more nothing less. My issue with stix is singular; dominating the board with his dysfunctional marriage and playing the fool.
Comments For This Post Are Closed
Secondary Sidebar
User Tools
FanPosts
Community blog posts and discussion.
Recommended FanPosts
Recent FanPosts
In Association With
Big League Stew - MLB - Yahoo! Sports
- Get a grip: Reds’ Todd Frazier loses bat, still connects for home run (Video)
- Eric Hosmer redefines the Baltimore chop with infield double (Video)
- College player makes impossible over the fence catch to save tournament game (Video)
- Security!: Camden Yards guard collides with Royals’ Mike Moustakas
- The Juice: Nelson Cruz matches career high eight RBIs, Pujols stays hot in Seattle
