The closer one night stand

The saves category is one of the more frustrating in fantasy baseball, involving a good deal of luck and favoring those with the time luxury or misfortune—however you want to look at it—of being able to immediately react at the first breaking news of a strained shoulder. Sometimes life takes precedence over fantasy baseball and you simply cannot compete with those people in the adding of newly anointed closers.

There still are ways to get some cheap saves that do not require you to be the first to jump on injury or some other news that results in a new reliever coming in the ninth. One way—the way I will go on to describe in this article—requires you to be in a league with daily roster updates and relatively deep rosters so if your league does not fit the description, I’m sorry, this strategy probably will not work well for you.

For those whose leagues apply, keep reading.

He has pitched in how many consecutive games?

The concept is simple: Keep track of closers that have pitched in consecutive games and consider adding the team’s setup man for one day, tomorrow’s game. If the same team is leading by a small margin in tomorrow’s game, they might not want to use their usual closer for a third or fourth consecutive night so you add the team’s setup man … and voila! The next game the usual set-up man pitches in the ninth instead, plays closer for a night and nets you an easy save.

Sometimes it works like charm, but often times things go awry.

Potential problems

I am not sure what percentage of MLB games include a save, but whatever that number is divided by two is the chance that the reliever you audition even has a chance of getting a save. Then, there is the chance the team uses its closer for a third straight night, or uses a different reliever as the fill-in closer.

Another problem that will occur more often in deeper leagues is that the potential fill-in closer (current set-up man) might be already owned. If that is the case, you can take a chance on a different reliever in that team’s bullpen or forget about it.

Overall this strategy has a low success rate, but the five (more or less) saves it can cheaply garner you over the course of a season may help you greatly in the standings. Some of you looking over your league standings can easily picture how much those extra saves could help right now.

Concluding thoughts

As I noted before, it helps if your league has deep rosters so that roster spots themselves are not as valuable and can be used on something relatively trivial like this strategy. Some people, however, seem to have a slight obsession with closer-potential middle relievers who are not getting saves. Instead of holding onto one of those Matt Thornton, Matt Guerrier-types, maybe the roster spot would be better utilized by rotating between relievers who fit the criteria above. It all depends on your team and league type.

A tool that surprisingly comes in handy for this strategy is the THT Sparkline Generator. Clicking on the link will show you how to set the sparklines to your custom settings so it shows the games a team won by four or fewer (set it to four or three) runs in the past week.

Teams like the Yankees and Rays with three consecutive red upticks probably have overused their closers in the past few days and those team’s setup men are good targets. Checking Mariano Rivera and J.P. Howell, the Sparklines were right, both closers made appearances the past three games and probably will not be used in a fourth even if it is a save opportunity.

With Monday night’s games now finished we see neither the Rays nor Yankees games had a save situation (Yankees were close), so last night would not have worked. If you continue to keep track of closer use throughout the season though, every once in a while you will get a surprise save and it will all be worth it.

Print Friendly
 Share on Facebook0Tweet about this on Twitter0Share on Google+0Share on Reddit0Email this to someone
« Previous: Don’t give up on batting average just yet
Next: And That Happened »


  1. Steve A said...

    This is an interesting strategy to pick off some saves.  I can see this working well in head-to-head leagues, especially.

    Thanks Paul!

  2. digglahhh said...

    The underlying point here is to have as few dead roster spots as you can. If you have a rotating spot or two, you want guys who are going to get into games and produce stats for you. Whether you want to chase the unlikely vulture save and tweak your strategy accordingly, or you want to just hedge your bets and get high quality per-inning production from middle relievers is a strategic decision for you to make.

    I like to rotate waiver wire bats to fill out my line-up on travel days and stream high K-per-inning middle relievers throughout the rest of the week. I’d rather have active MRs on Tues, Wed, Fri, Sat, Sun than a marginal bench bat who can be easily replaced off the wire. Because you’re frequently dropping and adding, you need a decent pool of available middle relievers, because at any given time your top choice may still be on waivers from the last time you dropped him.

    My strategy is just to go for best overall quality available and leave save opps or vulture wins to the serendipity of the fantasy gods. Last night, I’ve caught a both vulture wins and saves from the likes of Matt Guerrier, Dan Meyer, Rafael Soriano, Matt Thornton, etc. over the course of the year. And that’s great, but it’s the icing on the cake. Much of the best value you get from having a steady stream of MRs pitching innings for you is the dozens of extra innings at very low rate stats and high K-rate. For those of us who are often hesitant to spend premium picks on high-priced arms, getting 60+ of amalgemated middle relief work to a composite line of something like 2.25 ERA 1.15 WHIP and 60Ks is a nice way to help stabilize your staff. You can get lucky and pick up a half dozen or more wins and/or saves in the process, but I don’t know if it’s reasonable to really “chase” them. What I do know is Melvin Mora sitting on your bench isn’t going to get you any wins, saves, strikeouts, or 1-2-3 innings.

  3. digglahhh said...

    Sorry, meant to say that last night I caught a win from Guerrier, and have caught wins and/or saves from those other guys too, throughout the season.

  4. John Burnson said...

    This is a cool idea! I checked my database of all games from 2006-09. I queried for all streaks of three Saves on consecutive calendar days.

    On those occasions when a team used the same guy to close the first two games, they used a different guy on the third night 1/3 of the time.

    The downside is that, from 2006-09, there were only 156 such streaks. On the other hand, that still rates to about 1.5 such streaks per week. Optimally, then, you could acquire about half a save per week (1/3 of 1.5) or a dozen saves for the season. This assumes that you correctly peg the fill-in, and that his team actually wins.

  5. Paul Singman said...

    I like your strategy of filling out your hitting lineup on common travel days and then auditioning middle relievers on the other days, hoping to maybe get a surprise save or win… But why leave the vulture saves “to the serendipity of the fantasy gods” when you can do a little research, find out if there are any setup men with overworked closers to add, and increase your chances of getting the save.

    Thank you for finding out the actual numbers. When you say they used a different guy 1/3 of the time, is this 1/3 in the next game regardless of situation, or only 1/3 when another save situation occurred?

    @MadMax and Steve
    Glad to help!

  6. digglahhh said...

    Sometimes I do, and sometimes I don’t. It depends on both my available time, and motivation.

    It’s kind of an ROI thing. If you’re doing this regularly anyway (rotating roster spot), then there will be days when no such opportunity exists. So, that will be wasted effort. When you do identify such an opportunity, you only score (a save) when a)that team wins, b)there’s a save situation, c)the team chooses not to use their closer for another consecutive day. You need the stars to align, and while research can increase the likelihood thereof, it’s questionable whether the results are actually worth the extra effort.

    An interesting experiment would be if one team constantly jumped from most likely save/win vulture of the night on a daily basis, and another simply just rostered a high quality, non-drafted middle reliever for the whole season (Matt Thornton, say). Then compare the composite stats of the rotating spot to Thornton’s and see who actually got more value. I’m not sure what the answer would be.

    Some years, there’s a better candidate for the ideal perma-middle reliever than Thornton or Guerrier too. (Say you started out the season with this strategy but chose Rafael Soriano or Andrew Bailey instead – you’d be sitting pretty. Had you played the wheel, you probably would have owned and dropped both of those guys and would be kicking yourself).

    For a few years, (maybe ‘05-07) I was fond of drafting one fewer starter than I really wanted and blowing late picks on both Scot Shields and Scott Linebrink, and thinking of the tandem as a combined starter. Sure, it thinned my roster a bit, but combined they were probably good for 12-14Ws, 8-12SVs, 140Ks, and a low 2s ERA, and 1.00-ish WHIP over 140 innings. Wouldn’t you trade your 23rd and 24th picks for one pitcher with that line?… (Of course, the real issue with this is the opportunity cost of catching a breakout star with those late picks – but that’s almost impossible to quantify).

    So, getting back to question, I question 1)whether the effort of the research is just like buying 4 lottery tickets instead of one (doesn’t improve your chances in any manner worth the extra expenditure), and 2)whether you may be better off just holding on to the elite middle reliever.

    For question #2, I assume it’s all subject to the league, available players, and team needs. I’m just saying, last year Jose Arrendono was likely more productive than the carousel would have been – and there’s always a few guys who are like that every year.

    Here’s another thought at the end of this marathon post, how about just diving a pitcher’s number of decisions + saves by his IP, thus determining pitchers who are most likely to get a vulture save or win, overall.

  7. Paul Singman said...

    Digglahh, I agree that for some people the time and effort might not be worth the minimal possible reward, but to the most hardcore fantasy ballers using the sparkline generator to get a few cheap saves a year may seem more than worth it.

    Regarding your suggested experiment, the results would be irrelevant since it is all about individual situations that should not be generalized.

    Both strategies have their risks and rewards: the high quality middle reliever one is more all-or-nothing and the rotating relievers strategy reward is more of a consistent-little-something.

    My personal recommendation is to not chose one strategy or the other but to combine them. If you come across a middle reliever you really like and you think is in the perfect situation to get saves soon, then hold onto him. However, if no great options are available, see if you can get a cheap save or two until the next guy worth holding comes along.

    A lot of people are proponents of the draft the best middle relievers late in the draft strategy, and I am as well to an extent. My main criticism of it, though, is that often the best MR of one year is simply average or even worse the next. I’m thinking of guys like Scot Shields, Rafael Betancourt, and Grant Balfour. I looked into this in this article almost a year ago.

    Last thing, regarding your final statement, I would imagine players that have the best ratios at one point in the season will not continue to lead the pack over the rest of the season because of random chance alone. Vulture saves and wins are the result of lucky timing, which is why the strategy is suggest in this article is valuable—because it increases your chances of success.

  8. digglahhh said...

    That is indeed the difficulty with the drafting of MRs. They are, by definition, the least talented group of pitchers in the game. Were they better, they’d either be starting or closing. Very few retain high levels of performance year to year.(There are many reasons for this, which I don’t need to explain to this audience) In their heyday, Shields and Linebrink were consisstently stellar year to year, clear heirs to the closer job in the case of injury, and involved in a high percentage of “decisions” (brought in very often, and in close games). Very few middle relievers were considered “draft worthy” going into this year (excluding closer speculating): Putz, Soriano, Marmol, maybe Zumaya? I’m probably forgetting somebody. Linebrink and Shields were safe bets in their primes, that’s my only point there.

    Basically, I only trust “brand name” middle relievers if I’m adopting the stick-with-one-guy approach.

    And, I wholeheartedly agree, that whether you want to use one elite guy or the carousel is an individual, strategic decision.
    The risk-reward of high roster turnover strategies is inherent. In a nutshell, the upsides are that you get more chances at production, and if you are smart you can maximize those chances and tailor them to your specific needs at different times. The downsides are that you can also increase the risk of negative production (your reliever gives up 3 earned over two –thirds, or you pick-up bat goes 0-5), and that you will likely look back and see that you at one point owned, and then dropped, many guys who had great years because you didn’t have the patience to see them develop, or you didn’t trust the production enough to protect that player from the chopping block on his first day off.

    It takes both skill and luck to reap the highest rewards from the carousel approach!

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>