LaTroy Hawkins and Dave Cameron’s Radical Proposal
Dave Cameron made a radical proposal today aimed at teams thinking of moving away from the standard five-man rotation. He suggested that they essentially blur the lines of starters and relievers and simply assign an even number of innings (or batters faced) to each pitcher in the pen. “Each pitcher will be asked to face 10 batters per game, which translates to about 38 pitches apiece.”
Cameron advocates that teams target two-way players like Micah Owings, who are decent at handling a bat and also can throw average innings, and swingmen like Alfredo Aceves — mopup guys capable of taking a fair number of innings fairly frequently. That’s actually much more akin to the way that bullpens used to function in the days before hyperspecialization, except that in those days the majority of the important innings went to a “closer” like Roy Face or Bruce Sutter or Rollie Fingers, who pitched a lot more innings but had a lot fewer appearances than modern closers and setup men.
Cameron’s strategy makes pitchers more fungible. But teams would still want reliability. The best pitchers in such a system would be the ones who throw a lot of innings every year without fail. There aren’t many of those around: there are only two active pitchers in the top 40 on the all-time appearances list. One of them, Mariano Rivera, is out for the year and maybe for good. But LaTroy Hawkins is still going strong.
He’s busy climbing up the list of the most longevitous relievers in baseball history. His 839 appearances make him 38th of all time, and his 741 relief appearances make him 35th of all time. If he pitches another two seasons, he’ll be the age that Arthur Rhodes was last year, and he has a decent chance of passing Rhodes (839), Doug Jones (842), and getting within spitting distance of Billy Wagner (853), who is 27th on the list of most relief appearances. (No one’s catching #1 Jesse Orosco, who has 1248. Number two is Mike-not-Giancarlo-Stanton, who has 1177.)
The thing that makes Hawkins so amazing is the reliability of his arm. He wasn’t much of a starter and he wasn’t much of a closer, but from the moment that the Twins made him a full-time bullpen arm in 2000, he pitched ten straight years with at least 50 innings in relief. He’s just the 36th pitcher in baseball history to do so.
It’s a relatively recent phenomenon. The earliest reliever in baseball history with ten consecutive seasons of at least 50 innings was, rather naturally, Hoyt Wilhelm, who may be the best relief pitcher of all time; he had 16 such seasons. (Wilhelm is one of three Hall of Famers in the list, along with Sutter and Fingers. Goose Gossage and Dennis Eckersley didn’t manage ten in a row as relievers.)
Wilhelm’s total was only exceeded by that of Lindy McDaniel, who had 18 such seasons, 17 of them in a row from 1959 to 1975. Here is the complete list:
| first year in streak |
last year in streak |
# in a row |
|
|---|---|---|---|
| Hoyt Wilhelm | 1961 | 1970 | 10 |
| Ron Perranoski | 1961 | 1971 | 11 |
| Don McMahon | 1958 | 1972 | 15 |
| Ted Abernathy | 1963 | 1972 | 10 |
| Jack Aker | 1965 | 1974 | 10 |
| Lindy McDaniel | 1959 | 1975 | 17 |
| Clay Carroll | 1966 | 1977 | 12 |
| Darold Knowles | 1969 | 1978 | 10 |
| Paul Lindblad | 1968 | 1978 | 11 |
| Sparky Lyle | 1968 | 1981 | 14 |
| Dave LaRoche | 1971 | 1980 | 10 |
| Tom Burgmeier | 1974 | 1983 | 10 |
| Tug McGraw | 1969 | 1980 | 12 |
| Gary Lavelle | 1975 | 1985 | 11 |
| Bruce Sutter | 1976 | 1985 | 10 |
| Tippy Martinez | 1976 | 1985 | 10 |
| Rollie Fingers | 1971 | 1982 | 12 |
| Gene Garber | 1973 | 1987 | 15 |
| Kent Tekulve | 1975 | 1989 | 15 |
| Greg Minton | 1979 | 1989 | 11 |
| Dave Smith | 1980 | 1990 | 11 |
| Jeff Reardon | 1980 | 1993 | 14 |
| Lee Smith | 1981 | 1993 | 13 |
| Eric Plunk | 1988 | 1999 | 12 |
| Robb Nen | 1993 | 2002 | 10 |
| Steve Reed | 1993 | 2004 | 12 |
| Roberto Hernandez | 1995 | 2006 | 12 |
| Steve Kline | 1997 | 2006 | 10 |
| Mike Timlin | 1996 | 2007 | 12 |
| Todd Jones | 1994 | 2007 | 14 |
| Jose Mesa | 1994 | 2007 | 14 |
| Alan Embree | 1998 | 2008 | 11 |
| Trevor Hoffman | 1993 | 2002 | 10 |
| David Weathers | 1998 | 2009 | 12 |
| Guillermo Mota | 2002 | 2011 | 10 |
| LaTroy Hawkins | 2000 | 2009 | 10 |
Running down that list, you can see many of the most reliable and most beloved relievers of the last half-century, if not necessarily the best-paid. (Mariano Rivera just barely missed the list, because he only pitched 46 innings in 2002.) If a team were to implement Dave’s proposal, they would want guys like the ones on this list: multi-inning closers like Garber, rubber-armed middle relievers like Weathers, year-in-year-outers like McDaniel.
Trevor Hoffman was viewed as a future Hall of Famer in his day, and his ilk of relief ace would be just as scarce under Cameron’s as they are now. But bullpen glue guys like Timlin, Kline, Reed, and Embree are often undervalued, because they are both good performers and extraordinarily healthy. Obviously, it’s easier to determine a proclivity for good health in hindsight. But it’s one of the most important determinants of a player’s value.
Pitchers like Hawkins are underrated in the four-starter and five-starter systems that they’ve pitched in. In Cameron’s no-starter system, they will be just as valuable, if not more so.
How will you ever get starting pitchers to buy into this idea when wins are so important to them?
I just cannot see this experiment being successful.
It’s a clown idea bro
You wouldn’t do it to highly-paid guys, who have a serious economic incentive to resist — besides, as Dave Cameron said, if you have Justin Verlander, you’re not going to cut him back to nine hitters per start.
My guess is that do this with relatively low-paid guys, probably mostly team-controlled guys, or swingmen/longmen who typically only sign one- or two-year deals. Guys like Aceves or Owings.
I just don’t think there are enough two way players/swingman like Owings and Aceves for this to work.
Swingmen are just 5th/6th starters and Quad-A pitchers. There’s plenty of them.
Exactly. They are a dime a dozen. They just don’t fit in the current paradigm.
Dave also suggested that a team needs to find a new way to compensate pitchers in order to get them to buy into this.
So we get a bunch of guys that aren’t very good and try to limit how many each one of these bad pitchers pitch . At the end of the day you still need to pitch 9 innings
If the Nats were to shut down or severely limit Strasburg’s innings in September, you could see them calling up a few additional relievers on their 40 man roster and doing this with Detwiler, Stammen, and Gorzellany, or perhaps Strasburg along with those 4. This could keep Strasburg active and perhaps give him a longer start in each playoff series should they get that far.
I think this would work better if you went sort of half and half. Most teams have at least one or two good starters that they would like to see pitch over 200 innings. So have those guys in the rotation like normal, but fill the other three spots with long reliever types as suggested here. That way you could have your Verlander of Sabathia and still have your improved “bullpen days” over crappy starters.
This is a very good idea. Some existing teams should definitely try it, especially in the NL where the team could gain from pinch-hitters instead of pitchers batting.
Yea, for a team like Seattle this year, where you have 1 good starter (Felix), 2 hit-or-miss innings eaters (Vargas and Millwood), 2 bad starters (whoever) and a seemingly endless supply of decent relievers, I think a split system would create significantly better results than they’re getting right now. They wouldn’t even need to rebuild the roster to fit the system.
I can imagine Dave might’ve come up with this idea in the first place watching a Hector Noesi start, and thinking it would’ve been better to just run out the bullpen for the full nine.
If you limit players to this type of role does that at all diminish their chances of developing into a true ace? Would teams still be giving their top pitching prospects the opportunity to throw “Verlander” innings knowing that their is potential there?
1) Developing aces: True, this system doesn’t. It makes good pitchers out of mediocre pitchers. Rivera, Paplebon, and other closers are mediocre or failed starting pitchers. There are few aces. They are all freaks. And they want and have earned a big part of your budget at the point where they are due to break down.
2) Pitchers buying in: It’s an easy sell to relief pitchers, obviously, and if your starting pitching is crappy enough to risk breaking baseball tradition, who cares what they think? True, breaking tradition is like a pitcher stepping on a foul line, and is in their little heads like a tick on deer. But they will be competing with established relief pitchers for important roles, and sulking isn’t going to help them shine. When a guy with a 4.50 ERA becomes a guy with a 2.75, and his team is competing every night, he’ll get over himself. As to pitchers avoiding the organization because there is no opportunity to be a diva, let us remember that it also may be seen as the fastest path to the big leagues.
I have never heard a professional position player contradict the assertion that pitchers are the dumbest players on the field. It is the job of the manager to outsmart them. If the manager is the second dumbest guy on the field, that might not happen.
Alex, maybe you can get Dave to actually respond to this. He stated as a comment in the previous thread:
“Early game performance doesn’t predict pitcher performance any better than a pitcher’s true talent level. Just because a guy gives up three runs in an inning doesn’t mean he doesn’t have it, or that he’s going to keep struggling if you leave him in.”
To which I asked: “Please show me your research on that. I have seen this stated ad nauseum, only to be handed work with an extreme selection bias each time.”
Not that I think the idea can’t succeed…but it is built on some rather large assumptions that require empirical evidence.
I haven’t seen the research but I am certainly not as up on a lot of that data as Cameron is. I think your best bet to get a direct response from Cameron is to comment on the story or ask the question in a live chat.
I did, to no response. One would think such a crucial assumption, when made, would be sourced.
I don’t have the database skills to pull it off, but I imagine you could do a study with Retrosheet data, looking for pitchers who had a three-run inning in the first inning or two of the game, and seeing how they pitched in the following innings, and compare that to their averages for those innings.
Still ripe with selection bias, as it leaves in a critical assumption that those taken out and those left in had the same ‘stuff’ that day…which can be controlled with pitch f/x. It also requires proper scaling, as those that would be left in after surrendering three would likely be a different set of pitchers from those removed.
Never seen anything like that completed, and without it completed, the theory is nothing but poorly formed conjecture.
I actually don’t think the selection bias — being limited to pitchers who would be good enough for their managers to leave them in after they gave up 3 runs — is that big a problem. The most important thing is to compare apples to apples. If Roy Halladay gives up 3 in the first, does that mean that he “doesn’t have it” that day? Does he have an increased likelihood of giving up more runs in the second, or not?
This doesn’t necessarily answer the entire question, but it answers part of it.
LaRussa tried this already and abandoned it in like two weeks. Stop doing vudeo game contortions to plug in players like robots or numbers in a formula. The solution for a team that needs to get more out of its pitchers is to develop better pitchers to begin with. Honestly, some organizations just need to completely overhaul their conditioning programs, instruction, scouting, free agent spending and so on, and make the hundreds of little tweaks to build a quality franchise top to bottom. This notion that you can turn chicken scratch into chicken salad simply by shuffling a couple players’ roles around is just taken to absurd extremes sometimes. It becomes the sabermetric analog to the traditionalist’s “well, just go spend on a big bat”.
Sounds as if this would work for a team trying to survive a season sans strong pitching — a la the Colorado Rockies — but this is no way to win a division title.