Closers - drafting strategies

[viper]

For a couple of years I feel I have bombarded with the idea that I should not expend either high picks in a straight draft or spend any real money in an auction draft. The generalized statements are you can ALWAYS find saves.

Poppycock !!

In today's world with unlimited access, smarter owners and more leagues going towards FAAB, new closers are not that easily obtainable. In the older ESPN days, the person who lived at their computer could be first to get a newly anointed closer. It was survival of the fastest typing finger. Today when your FAAB day rolls around, everyone knows who is available and blind bidding wars occur for every new closer. On top of that, those closers-in-waiting are often times drafted and stashed.

Todd's Paplebon Plan identifies this new reality and provides a blueprint for success - at least as long as today's current closer strategies are continued.

The only constant to closers and drafting is that all closers will be drafted. A near constant is that all superlikely-to-be closers will be drafted.

In mixed 15 team leagues, you either plan to get:
3 closers and get a high score by brute force but this does hurt hitting some
2 closers and shoot for a middle score with the better the closers, the higher middle you get
1 closer and try and beat all those with zero and those with just one like you - this allow for better hitting
0 closers and try to get a few points through FAAB but spend more on hitting and SPs.
My plan is almost always have two closers.

In mixed 12 team league, the options are the same but now at least six team will have three closers - more likely seven or eight. If seven, this means that two closers will get at most five category points. The more teams that have one or zero, the range for two closers could be very small and makes this number less appealing.
It's been a long time since I played in a 2-team mixed, no clue what my plan would be.

In 10 to 13 team single universe leagues, you either get two, one or zero. Two closers gives you an above score at the expense of hitting. One closer put you in the bigger battle with the most teams. With just one, you need a closer who gets the most saves - one of the top tier closers regardless of cost. With zero closers, you get a low score but save picks and/or dollars for better hitting and pitching.
I generally end up with two closers - one quality and the other 3rd tier

Thoughts, comments, your normal plans?

[Hambowen]

I usually stay away from the top guys and take risks. I try for a $15-20 guy and then go for a sub $10 guy that I hope/think could earn $15-20. In leagues with benches I then stash potentials.

At worst assuming health/role does not change you will be in the lower middle as far as points go. If things work out well you can be in the upper middle. So anywhere between 4-9 points in saves.

This changes a lot however based on value. If the top guys are undervalued in the draft then I will grab one of them. So as usual flow of the draft can change any strategy.

[rotodog]

I play Keeper leagues only so my closer strategy is much different. As i state everywhere, you need to know your league well to put together a plan for all players, not just closers.

For closers, I hate to pay at auction for them. In a keeper league I may own a cheap closer already or maybe two.... But I can always pay for "next in line types" on the cheap or get them in trade by dumpers.

I can tell you this. Having one closer is useless in most leagues unless you can add to that total easily via trade/FA or you have 2-3 Setup guys on teams with shaky closers on the bench or active. One year I owned Baez and he was my only closer in an AL only league. he got 45 saves that year..Not too shabby huh? I still got only 2 points in saves and only beat a guy that punted saves because nobody traded closers in that league and many of the up and comers were on rosters already. I essentailly wasted my inflated 26 bucks on Baez when I could have spent it elsewhere...Wasted money.

If I own a cheap closer in a keeper league, I will add one low tier option and a couple "next in line types" and pray...A trade or two and I will always end up middle pack and higher...

I just hate paying for closers on draft day and feel that money is spent elsewhere. I also didnt play in standard K leagues either and owned a lot of relievers because K's were replaced with another ratio category . I just hate paying for pitchers at all generally and feel i am good at finding value and talent cheaply....Its worked well for me and I have cashed many checks in October although it is never about the money....

But if closers are going super cheap, I will take it....

After all that, I guess I take what the draft gives me. My only strategy is that I hate paying for closers....

[cwk1963]

One of my leagues is a 16 team standard roster with only 3 bench spots - not a lot of room to stash people. Every year someone goes the zero closer route (I've tried it) and they never fail to finish at or near the bottom. Since it's a keeper league, I try to get a closer-in-waiting pretty cheap even if it's for the next year. I did this with both Heath Bell and even Kerry Wood when there were the faintest of rumors he would be tried as a closer. A lot of money is spent on closers. I try to get one in the off season by trade but that's not working well lately. I have Wood at $11 as a keeper and will try to get a top closer in the auction for $20. Less than 2 and, historically, you won't finish in the money.

[alleyoops]

I've struggled somewhat with closer strategy in my leagues, and I've varied the strategy through the years. I can't say that any one approach has proven better than another, in terms of my results. Here's what I've done most recently:

14 team mixed 5x5 keeper - bought one strong closer at a decent price (Nathan at 20), and two risky closers at a lower price. Added a couple as FAs during the season, which offset the risky ones who lost their roles. Wound up 4th in the league and near the top in SV.

12 team mixed 5x5 (k/9 instead of k) keeper 1st year of league - bought Wagner (yay) and Street (boo) at decent prices, aggressively pursued FA closers-to-be, grabbing a few. Finished out of the money, top half in SV.

12 team NL only 5x5 keeper - drafted no closers, traded for one early and added two as FAs. Dumped mid-season, but still finished pretty high in SV.

15 team mixed 5x5 draft non-keeper (FB.com staff league) - drafted a risky closer or two late, added a couple as FA speculations that worked out. Finished 5th and pretty high in SV.

Based on all of this, I think I'll likely take the spend little for saves approach this year, since I was pretty successful adding through trades and free agents. That is, unless the closer prices drop to a low enough level that I can't pass them up.

[50 Desert Eagles]

I usually end up in a 10 team NL league...auction or straight picks, I usually spend last 2 picks or $5 max on 2 FA type closers and after 2-3 weeks try and snag up someone losing there job or injured, there is ALWAYS atleast one or two....if not, then after 4-5 weeks, I offer up a SP I got(better then most because you got an extra 2-3 higher picks since you didn't get one or buy a RP in the draft) that has maybe a 3-0 or 4-0 record(such as a Randy Wolf or some rookie that everyone is deeming as the best ever) and look for the guy with 3-4 closers and few wins or poor ERA and offer up a deal.....more often then not, you can get a better then average RP this way....IF your in a good league that allows trades that is....I have a 12 team 5x5 mixed league drafting Monday night, I will try and post the results on here somewhere to help out everyone that does mocks and such....its a FB.com league, so I am not sure what to expect....either a bunch or knucleheads that take ALL Yankees or some wise guys that could be reading my thoughts

[shif6]

In one of my 5x5 auction keeper leagues (with generous rules on keeping players for long terms), owners pay 4x4 prices for closers (or keep them at high prices). So I rarely draft the few closers available in the draft unless a bargain falls in my lap. But I buy closers in waiting, work the wires and pray. As Viper observes the competition is fierce on the waiver wire. Over the years, however, I have accumulated cheap keepers. It would never occur to me in a redraft league to use that strategy. In a NFBC league I would leave the draft with at least two (and hopefully three) mid level closers and a couple of closers in waiting on the bench. As Viper suggests, closer strategy should depend on league size; it should depend on whether it is a keeper league or not; and it should depend on the rules for how long you can keep keepers.

[Todd Zola]

The aspect of the Papelbon not mentioned here is the usual plush layer of second tier closers, at least to me, smaller, while tier 3 is filled with some untested guys. In an auction, I can still cherry-pick the guys I like, in a draft, I am at the mercy of my fellow drafters.

Off the top of my head, in no particular order within the tier

TIER I (have job, done it before, elite skills)
Papelbon, Nathan, Rivera, Soria, K-Rod, Lidge

TIER 2 (have job, done it before, very good skills)
Cordero, Fuentes, Wood, Valverde, Capps, Jenks, Ryan

TIER 3 (have job, somewhat untested or skills only good or injury risk)
Bell, Corpas/Street, Marmol, Hoffman, Hanrahan, Devine/Ziegler, Gonzalez, Lindstrom, Broxton, Wilson, Francisco, Qualls, Perez

TIER 4 (job by default)
Lyon, Percival, Sherrill, Seattle????

In a draft, if you are near a turn, especially in something with more than 12 teams, you can miss out entirely on the first 2 tiers as either group can easily go between your consecutive picks, leaving you to throw darts at that 3rd tier. Will some succeed, putting up 2nd and 1st tier numbers? Yup. Will they all? Probably not.

[uharchmajor]

TIER I (have job, done it before, elite skills)
Papelbon, Nathan, Rivera, Soria, K-Rod, Lidge

TIER 2 (have job, done it before, very good skills)
Cordero, Fuentes, Wood, Valverde, Capps, Jenks, Ryan

TIER 3 (have job, somewhat untested or skills only good or injury risk)
Bell, Corpas/Street, Marmol, Hoffman, Hanrahan, Devine/Ziegler, Gonzalez, Lindstrom, Broxton, Wilson, Francisco, Qualls, Perez

TIER 4 (job by default)
Lyon, Percival, Sherrill, Seattle????

I guess this also brings up a question on how you do the projections? If closers are as risky as everybody knows they are, how do you project TIER 3 closer? Without giving exact info away with regards to MB projections:
TIER 1 player
39 saves, 84 K, 2.73 ERA, 1.21 WHIP

TIER 2 player
36 saves, 70 K, 3.00 ERA, 1.26 WHIP

TIER 3 player A
34 saves, 95 K, 3.15 ERA, 1.13 WHIP

TIER 3 player B
34 saves, 59 K, 3.29 ERA, 1.36 WHIP

TIER 3 player C
31 saves, 80 K, 3.55 ERA, 1.31 WHIP


I can understand creating tiers because, when drafting a TIER 1, you know what you will get and the further down the tiers you get the more likely the projections are off. (apparent with the increase in ERA - more runs could lead to more blown saves, could lead to being pulled from the closer role)

[Todd Zola]

The projection tiers and my anecdotal tiers are two different entities. Skills/ratio/performance projections are based off of past performance. We have an essay that will hopefully be posted by the end of the day that discusses how we project saves.

My anecdotal tiers are a bit different 1-elite skills, job secure with little injury risk, history of success, 2-same as #1, but skills not quite the same, 3 - a chink in the armor somewhere, could have great skills but a newbie in the role, could have great skills but be injury prone, etc. 4 - role by default

A highly skilled closer is in my subjective Tier 3 likely because he does not have a history of doing the job or is injury prone, but probably the former.

In an auction, I still favor aiming for those in Tier 3 that I trust, assuming I can't get someone in Tier 2 at value. I will be sharing those names in the Platinum material down the line.

In a draft, I am at the mercy of my fellow drafters. I could get caught in the wrong side of a run and miss out on Tiers 1 and 2, then be forced to get two or even three guys from Tier 3. Again, I am now at the mercy of my competitors, as they are likely to favor the same guys in Tier 3 that I do, so I either have to jump the run or hope one falls. The chance I get two that I like is slim, unless I jump the run, which I will do, assuming I did not secure a Tier 1 or 2 guy earlier.

[alleyoops]

I think the implied question above was - the projected value of closer A in Tier 3 would be the same, or maybe higher than the projected value of the closer in Tier 1 - is that somehow accounted for in the projected values? I think the answer is probably that it is not.

So, we need some way to know that we'd much rather have the Tier 1 guy than the Tier 3 guy in a draft or auction. I guess the tiers themselves are this guide. Well that, and our knowledge of the players' histories and current situations.

I think another site assigns a "reliability" factor to each player - basically a measure of how likely the player is to hit the projected numbers. Do you have plans to do something similar?

[Todd Zola]

Again, the projections and my anecdotal tiers are completely different things. We try to preach not to take any of our projections verbatim, which is why we are in process of supplying the detailed projection and valuation methodology.

The primary differentiating factor is my delineation of "done it before" versus "hasn't done it before", really, that's it. So a first time closer with an outstanding history of set-up (Marmol, Broxton, Bell, Qualls to name a few) are going to have very good stats projections and the number of saves commensurate with the team they are on. When looking at projection tiers, they will be higher.

As for risk assignment, it is something Gary and I have talked about, but perhaps in a different manner. Quite frankly, I think the risk assignments are nice, but are barely above common sense. I just don't think many people need a pay service to tell you a guy is young with limited track record or injury prone. Gary and I are talking about how to attach error bares to skills projections, which in turn can be extended to the end-result, and have the final value be expressed via a range.

Some people are not going to like this, but our style is not to have you blindly accept our projections and associated value, but rather understand where it came from and make your own subjective opinion with respect to its validity and how you want to utilize that projection in your plans.

Fortunately, closers are really the only position/classification that this has such a big subjective influence. All we can do is objectively project the stats, make subjective changes as necessary, and subjectively determine the percentage of team wins will be saved by that pitcher.

One of my personal strategies is to be wary of closers with a limited track record -- that's what I am talking about in the Papelbon plan. How can we reflect that in a projection, reduce the number of saves? Is that fair?

What if my personal drafting strategy was to only select players under 30? Do I need to put an age tax on everyone over 30, dropping them in the rankings?

The best we can do is offer the reasoning behind the projections and valuations and share our personal subjective feelings and thoughts when it comes to how to utilize the projections.

[alleyoops]

As usual, a very good, and complete answer.

It does seem a bit inconsistent that injury history of a player has a large influence on the projections (in AB or IP), but "role risk" (for lack of a better term) has little or none. So the projection on Qualls, for example, assumes he'll be the closer for much of he season, even though there is a fair amount of risk of that not being the case. But the projection for Smoltz assumes he'll only pitch half a season.

I'm not saying that these are wrong, or that there's a better way to do it. Just that the approach to the two somewhat similar issues is different. In Smoltz' case, the projection is a big, red flag that says bid at your own risk. In Qualls' case, there's no flag, but probably the message should be the same.

Before you say it, yes, I should know that they both have risk, if I have any clue at all about baseball. And I did get the message about the projections being tools to be used in the context of thought and strategy, not THE absolute.

[Todd Zola]

Actually the Qualls answer is fairly simple in that I believe he will indeed keep the job, and that is reflected in the projection, as opposed to hedging and giving more saves to Rauch or Pena.

I did hedge a bit with St. Louis, Oakland, Tampa and Colorado.

Remember, I did say that there were some Tier 3 guys that I plan on targeting in auctions, and I will indeed include this in a Platinum essay.

[uharchmajor]

Remember, I did say that there were some Tier 3 guys that I plan on targeting in auctions, and I will indeed include this in a Platinum essay.

I'm looking forward to it, as after looking at different projections, I've found a few Tier 3 guys that could have some value later in my points league draft. Luckily CBS seems to have them projected much less then everybody else and during the draft, the players are listed by CBS projections.

[Guest]

There's a lot of gold in here which only partially has to do with saves. So thumbs up to all in here, seriously.

I'll just hit as many points as I remember.

1. As for reliability score, yes that is done by a very well regarded (deserved rep) site, its' something we've discussed a bunch, we've thrown a few methods at it and the worry we've had is that it can be easily abused/misused by a user - and maybe Todd and I worry about this to a fault, but we're so concerned about our product being transparent and properly applied that something like this maybe we're too apprehensive about.

In an ideal world, reliability would represent for us the likelihood of a projection being accurate. But this brings out several questions: A) Do we want to "penalize" guys with a solid shot at over-performing the projection by lowering their reliability for that reason? B) Does a crappy player projected to earn a buck because he's always earned a buck get a high reliability score because he's consistently crappy or do we have to weight it somehow by their ability.

Just things to consider.

2. Tier C/D closers. It's a fair point to raise that we perhaps don't penalize these guys enough, and I think that maybe only in closers do we really have this significant hedge - which is "assumes they pitch as they have, and assumes they keep the job" -generally, in a few cases we really do the hedge quantitatively. CJ Wilson got the closer job last year, had 2 years of great peripherals preceding 2008, and proceeded to walk the entire park in the closer role. (yes there were other issues). The real question is what to do there - we don't really like the idea of haphazardly cutting their skillset down to take the risk into account, and if the guy pitches as his skills previously allowed he would keep a full years job, its a quandry maybe best served qualitatively.

(of course, by doing this we de-weight the value of top closers by making it seem like there are other similar options available later, so depending on your league and the availabilty of saves on the wire, you may need to adapt strategically a bit).

3. The comment about further down the tiers, the bigger the risk of a projection being off is an interesting one. I read it at noon or so and contemplated it through a 5 year olds birthday party and dinner with my inlaws and watching BC lose to Wake this afternoon.

And I suppose where I'd leave it is that the art of capturing a save and pitching as a closer is as close as I can get to comparing a player in the minors moving to the majors. That is, we have a method for evaluating them and translating their stats that actually is pretty damn good (as said other places its very similar to MLE), but until they do it in the majors, we discount them because well, actually doing it in front of 35K fans is just a little different. Same with closers. I wish I could give a more quantitative answer but I suppose we don't really have a great way of translating pitching as a closer vs pitching as a MR. As a closer, you finish your innings for the most part, there isn't a harder throwing guy behind you to clean up your mess (though alternatively you usually can't blow up for 5 ER in an inning because 2-3 loses you the game), there's the pressure of blowing games for your team, etc. There probably should be a mathematical calc for this to somehow regress skills displayed in the MR role to translate them to the CL role, but alas, we don't.

[aburt19]

TIER 3 (have job, somewhat untested or skills only good or injury risk)
Bell, Corpas/Street, Marmol, Hoffman, Hanrahan, Devine/Ziegler, Gonzalez, Lindstrom, Broxton, Wilson, Francisco, Qualls, Perez


In a draft, if you are near a turn, especially in something with more than 12 teams, you can miss out entirely on the first 2 tiers as either group can easily go between your consecutive picks, leaving you to throw darts at that 3rd tier. Will some succeed, putting up 2nd and 1st tier numbers? Yup. Will they all? Probably not.

I agree that tier three closers will vary widely, with some putting up upper tier numbers and other bombing out.

Last year in a 15 team draft, I drafted Soria, who probably would have been an edge of tier 2/tier 3 closer and
J. Borowski, who I thought was a tier 3 closer and probably was actually a tier 4 closer. One huge hit and one huge
miss.

I will probably do the same thing again. I can hope that both hit and even if it doesn't, I ended up with a 6 in the
saves category. Given the cost of drafting a tier 1/tier 2 closer early, I'd rather wait and take the chance on tier 3
closers.

[steve9781]

I have and always will suscribe to the don't pay for saves. I do more drafts than auctions but the strategy while different does remain the same. I'll give my opinion on both.

Auction - I'll always try to grab 2 tier 3/4 closers for a combined $20 tops. If it were me, I'd target Bell & Qualls as my first closer, and then grab someone like a Devine / Ray / Lyon / Perez. There are studies done every year that show how many saves "come into a league" each and every year as opposed to new hitting like HR/SB. The numbers are never really close. I'll rarely ever finish near the top, but depending on how other owners are valuing saves I'll always pick up points over at least 2-3 teams. While I highly respect Todds opinion, I think his Paplebon strategy is nuts. He spends $30 on Papelbon getting 40 saves, I'll draft 2-3 $10 closers bank on 1 or 2 panning out, use the wire and finish ahead. It's the not putting all your eggs in 1 basket theory. IF Papplebon goes down, you've wasted $30 your never getting back. I understand every player is a risk for getting hurt, but why invest $$ in a position with as much volativity as closer

Draft - I almost never spend a pick in the first 10 rounds on a closer. I know each draft is different, but I smile everytime I see someone roster tier 1 closers in the early rounds of a draft. For me this goes for all pitching. You'll NEVER see a Santana / Sabathia / Paplebon / Nathan on my roster. For my money you lose so much hitting value, that it makes catching up EXTREMLY difficult. While your drafting P, I'm selecting stud OF, by the time the later rounds come around your rostering F.Lewis & D.Dejesus as your 4th OF, where I'm grabbing Greinke/Meche Cain or Garza. The stat difference between the pitchers I draft vs the OF your forced to draft is HUGE. Breakout pitching happens every year, there are 10 "breakout pitchers" for every McClouth.

Just my 2 cents.

[viper]

steve - in the straight draft NFBC would you follow the same strategy? Realize (1) there is no trading allowed (2) most of the obvious to semi-obvious closers-in-waiting get drafted, usually before their time, (3) most closers are gone after round 15 and (4) FAAB is used so all unexpected replacement closers are highly sought after and therefore expensive. You are, in essence, punting saves. I'm not saying this is right por wrong but stating the net effect of your strategy. In the main event where individual league payouts are moderately poor in order the fund the grand prizes, dumping a category eliminates any overall chances. In satellite leagues, I haven't heard of teams winning with a "1" or "2" in saves but I suspect it has happened.

I don't think Todd is proposing this strategy in an auction draft but in a straight draft but i could be wrong.

My mock draft trials have been seeing what type of team I draft using an aggressive Papelbon plan. My first 11 rounds are hitters and two quality closer using the 5th & 7th rounds and even the 5th & 6th rounds. My observations have been that after the 4th round, there is a lull in the draft of hitters when top pitchers and some closer are drafted. This last about 2-3 rounds. Also abut this time, the incremental value of hitters from round to round is slight unlike the first four rounds where the value is a noticeable descending slope.

[steve9781]

steve - in the straight draft NFBC would you follow the same strategy? .

No I would not but I wouldn't vary it that much, it's the difference in picking a closer in the 7/8 round and not waiting until after 10. From what I have read, an NFBC draft is a lot different than leagues that allow trading. Where trading is allowed people for the most part feel more comfortable waiting on closers, and more are available later. I will say my leagues would be considered shallow, 10 team 5x5 mix 29 player rosters. The more teams in the league the more it needs to be adjusted.

[viper]

The Papelbon Plan is more designed to deep leagues. I will be doing one shallow league this season. My son has a 10-team mixed auction league. I'm not sure what my draft strategy will be but my current thought is a LIMA type staff and an aggressive Stars & Scrubs roster.

[cwk1963]

I have and always will suscribe to the don't pay for saves. I do more drafts than auctions but the strategy while different does remain the same. I'll give my opinion on both.

Auction - I'll always try to grab 2 tier 3/4 closers for a combined $20 tops. If it were me, I'd target Bell & Qualls as my first closer, and then grab someone like a Devine / Ray / Lyon / Perez. There are studies done every year that show how many saves "come into a league" each and every year as opposed to new hitting like HR/SB. The numbers are never really close. I'll rarely ever finish near the top, but depending on how other owners are valuing saves I'll always pick up points over at least 2-3 teams. While I highly respect Todds opinion, I think his Paplebon strategy is nuts. He spends $30 on Papelbon getting 40 saves, I'll draft 2-3 $10 closers bank on 1 or 2 panning out, use the wire and finish ahead. It's the not putting all your eggs in 1 basket theory. IF Papplebon goes down, you've wasted $30 your never getting back. I understand every player is a risk for getting hurt, but why invest $$ in a position with as much volativity as closer

Draft - I almost never spend a pick in the first 10 rounds on a closer. I know each draft is different, but I smile everytime I see someone roster tier 1 closers in the early rounds of a draft. For me this goes for all pitching. You'll NEVER see a Santana / Sabathia / Paplebon / Nathan on my roster. For my money you lose so much hitting value, that it makes catching up EXTREMLY difficult. While your drafting P, I'm selecting stud OF, by the time the later rounds come around your rostering F.Lewis & D.Dejesus as your 4th OF, where I'm grabbing Greinke/Meche Cain or Garza. The stat difference between the pitchers I draft vs the OF your forced to draft is HUGE. Breakout pitching happens every year, there are 10 "breakout pitchers" for every McClouth.

Just my 2 cents.

I've only played auction leagues for quite a number of years. Trying to get 2 tier 3/4 closers for $20 or less is almost impossible. Even closers-in-waiting (not necessarily those current 8th inning guys) go in the auction. The leagues are predominantly 14 and 16 team mixed. One of the leagues is FAAB so, as Viper says, those closers that come off the wire cost a bundle. The others are WW with reverse standings order so only a couple teams have a shot at the few saves to be had there. The trade market for saves is virtually non-existent. The reason is everyone knows the winning team has averaged 4th-5th place in saves so closers are held even by rebuilding teams. It's much easier to get a somewhat useful MI or OF than it is to get saves off the wire.