Random thoughts on one-trick ponies, tradeability, and ‘value’

Last month, fellow THT writer Derek Ambrosino wrote a couple of articles that spurred some debate in the comments (Approaching unconscious competence and One category roar, five category snore). I had some of my own thoughts on a few of the concepts and theories discussed and wanted to share them.

As some quick background, most of the discussion centered around the relative value of one-category players and players who are a little above replacement in all five categories but spectacular in none.

Drafting to trade

In the comment section of Derek A.’s first article, there was a little talk about drafting with the intent to trade. As a general rule of thumb, most fantasy analysts will warn against “drafting to trade.” Still, Derek A. argued that “the chances that somebody needs a 40-steal guy regardless of the rest of that player’s (lack of) skill set (or at least feels that they need such a player) is probably higher than the likelihood that somebody feels they need a Garrett Anderson.” I think most of us would agree with this statement (I certainly do), so the question then becomes: “Should this be a consideration when we’re initially drafting our players?” My answer to this question is a definite “yes.”

A couple years ago, I discussed one of my favorite mixed-league strategies: that of drafting high-upside players late in the draft. This seems to be a very popular strategy these days, and I think the principles of the strategy are very closely related to our discussion on the trade value of one-category studs.

When we discuss “high-upside” players, we generally think about young, toolsy players who have a higher probability of significantly outperforming their projections than an established, 30-something-year-old veteran does. However, I don’t believe that upside must be constrained to pure production. Why shouldn’t potential future trade value be incorporated in the “upside” bucket? After all, it all leads to the same goal: winning. Whether that win comes as a result of your 20th-round pick hitting like a third-rounder or as a result of you trading your 20th-round speedster for a top-notch SP shouldn’t matter one ounce.

Value is dynamic

Another facet of the comment section discussion dealt with whether or not we should be drafting one-category players in the first place. Reader Andrew P. talked about how he disliked the idea of forgoing “more valuable” players in order to achieve balance by taking a one-trick pony. Not to pick on Andrew, but I’m not so sure it’s as simple as that.

Value is a funny thing, in that it is never static. I think when a lot of people talk about draft-day value, they think of it in a vacuum—as a precise, static number—but this couldn’t be further from the case. Value is dynamic and is unique to every team at every pick.

If you print out a list of players and dollar values and take that to your draft, the truth of the matter is, those dollar values will only truly be accurate until the first pick of the draft is made. After players have been removed from the pool and/or added to your team, the value of the remaining players will change. It will change—even if only slightly—every single time a player is removed from the pool. Over the course of an entire draft, those values can change quite drastically, especially if you’ve overloaded on one category and are short in another.

Here’s an extreme example to ponder: Let’s say the ghost of Ricky Henderson (from his 130-steal season) is resurrected and you draft him. Then you add a 118-SB Lou Brock clone. With your next pick, your pre-draft cheat sheet may say that the 110-SB Vince Coleman impostor is the best player on the board, but in the context of your team (which now sports phantom Ricky and Lou), Vince Coleman is significantly less valuable.

Why? Because you don’t need those steals! You’ve got 250 under your belt already—quite possibly enough to win the category outright. So the value of the remaining steals in the player pool is essentially zero for your team. For some other team participating in the draft, Coleman will be very appealing. But for you, the relative value of steals is extremely low, in turn raising the relative value of all the other categories. And this happens every time a player is selected (just not as drastically)—supply changes, your team needs change, and thus, every player’s value changes.

I actually just had a similar situation play out in a mock draft I participated in for USA Today’s preseason magazine. I ended up with Adam Dunn, Russell Branyan and Chris Davis on the power side and Michael Bourn, Nyjer Morgan and Luis Castillo on the average/steals side. Once you take a player who will contribute heavily to HRs and RBIs but little to average and steals, the relative value of HRs and RBIs to your team decreases, and the relative value of average and steals increases.

I ended up doing a lot of “balancing” in this draft. I put balancing in quotes because it’s a word that often gets used without full understanding of what it means or why/when it should be done. I wasn’t just taking these one-trick ponies because I felt I needed “balance” (something I don’t feel is necessary just for the sake of it); I was taking them because their relative value was higher to my team because of its current makeup at that point in the draft.

The tradeability of different players

My last point today deals with the ability to trade a one-trick pony versus a guy who will help out a little bit in each category. Derek A. used Melky Cabrera as an example of the latter, so I’ll continue using him. He posited that Melky would be a lot harder to trade than, say, Elvis Andrus or Scott Podsednik who have much of their value tied up in one category. I absolutely agree, but I have a couple ideas of my own as to why this is the case.

The first is a pretty obvious one (and one Derek A. touched on briefly). Midseason, teams are often looking to trade for categories as opposed to players. If acquiring one player can catapult your team three or four points in the standings, that’s going to be a lot more appealing than acquiring a player who merely helps in acquiring three or four points across several categories. It’s a matter of leverage.

One other important consideration, though, is that in our 12-team mixed league example, players like Melky and Garrett Anderson are end-of-the-bench guys. They are the guys drafted in the last few rounds or taken off the waiver wire during the season. While opinions of the top players in the league rarely differ from owner-to-owner (I think we can all agree that Albert Pujols and Mark Teixeira and Jacoby Ellsbury are worthy of a pick in the first few rounds), opinions of the guys taken at the end of draft differ greatly. One owner’s late-round bargain is viewed as should-be-waiver-wire-fodder by another owner.

Feel free to compare rankings between different sites and you’ll see exactly what I mean. Guys appearing in spots Nos. 248, 249 and 250 on one site’s list might not appear at all on another’s. Last year, Baseball HQ loved Mike Jacobs; I hated him. I loved Nyjer Morgan; my own readers hated him smile.

A Hardball Times Update
Goodbye for now.

Opinions diverge greatly at the end of drafts, so if you’re the guy who owns Melky Cabrera, there’s a very good chance you like him more than anyone else in your league (this is true to an extent for all players you draft, but particularly among late-round selections). And if that’s the case, how are you going to get what you consider equal (or greater) value in a trade? While Melky’s worth is open to interpretation, Andrus’ ability to steal a base is much less so.

Further compounding this reality is something called the endowment effect—the tendency for people to overvalue or grow attached to what they already have. As a result of this, owners are going to be more likely to want to keep their own, unspectacular end-of-bench guys than to acquire yours.

Concluding thoughts

Just some stuff to think about. I realize these ideas weren’t completely related, but I think they were all worth putting out there. If anyone has any of their own thoughts or questions, feel free to comment.


11 Comments
Oldest
Newest Most Voted
Inline Feedbacks
View all comments
Andrew P
14 years ago

In your Henderson/Brock/Coleman example, Coleman’s value to your team wouldn’t be 0, it would be whatever his trade value is to the rest of the league.  To me, the question isn’t what value Coleman has to your team, it’s what value does a player you might acquire have.  Since each player taken alters the value available in the remaining player pool, taking Coleman would increase the value of the remaining SBs for your leaguemates.  Trading only 1 your speedsters, then, would allow your trade partner to jump up multiple places (from last to 2nd isn’t unfathomable in this hypothetical).  This would ostensibly give your trade piece even more trade value to the rest of your league. 

Any time we’re trading from a surplus, the most advantageous position to take isn’t to trade for equal value relative to your team, but to trade for equal value relative to your opponent’s team.  The dynamism of player value is something to be exploited through trade, rather than by surpassing Coleman for a lesser player.

Further, a benchwarmer typically isn’t going to net much value in a trade anyways.  If your benchwarmer is productive enough to plug into your starting lineup, however , you’ll get a better return by upgrading your starters and plugging in your bench player.

Essentially, what I’m saying is that not only is there a larger degree of variance in disparate player values at the lower-valued end of the spectrum, but that the converse is true as well; the higher valued players tend to have more static values, and will thus return more in trades.  Pujols and Hanley Ramirez are the consensus top 2, and there’s very little argument about it.  It’s obviously not a perfectly linear relationship, but as you go lower in player value, variance in valuation increases.  Therefore, it’s much easier for an opponent to agree on the trade value of a player that is universally valued highly than it is for them to agree on the trade value of guys on the fringe, especially given that fringe guys are much more likely to be dropped into the free agent pool.

This means that if you have confidence in your analysis of your bench players, it is more valuable to trade your players that are more highly regarded across the league (obviously for at least near-equal value, if not better), than it is to trade your bench guys.  The discrepancy in player valuations between you and your trading partner will be smaller, thus nullifying much of the endowment effect.

As such, the value of a player like Melky Cabrera isn’t in their trade value, but that they’re better replacements in your starting lineup than lesser-valued 1-trick ponies.  As Derek (Ambrosino) stated in one of the earlier pieces, the fringe across-the-board guys actually tend to be undervalued relative to the 1-trick ponies.

Biggy
14 years ago

Good points, Andrew P, especially about bench players.  A guy in our league this year dropped Kendry Morales early.  I picked him up as a bench player, behind Hawpe and BJ Upton.  When Kendry heated up, I could then trade the more established players for other established players that fit my needs better.

Another caveat about your later, high upside players is the tendecy to overrate them, not dropping them because you believe in them.  I had Fontenot this past year, and missed on Aaron Hill because I “knew” Fonty would turn it around.  Value is important, but don’t overvalue your later picks.

GTWMA
14 years ago

To extend Andrew’s point in a roto environment, the estimate of Coleman’s value to you during the draft is likely to be his estimated standing gains points based upon your league settings. 

Through the season (and even dynamically during the draft) a player’s worth to any team may vary, which is Derek’s point.  If I’m looking to trade him, how other team’s value him is the key. 

I agree that one-tricks offer greater trade value.  But, I would argue that they also involve greater risk.  A team that builds its SBs on one player loses a lot if that player is injured, while the balanced team, by diversifying in SBs, is less risky.

Eric
14 years ago

This discussion begins to lead into what I have always thought was an interesting strategy but one that I never had the courage to try. 

Namely, that you do in fact draft Vince Coleman, and then you continue to draft SBs throughout the draft in the hopes of creating a league-wide scarcity in a category.  In practice, you would probably need to focus on a combination of categories such as SB/R. 

The aim would be to create a “monopoly” on a statistical category.  Via this monopoly, you would leverage the relative scarcity of the category in your league to make trades for stats that were correspondingly under-valued due to their plenty.  Basically, statistical arbitrage if that makes sense.  I can expand on the fundamentals of this concept if people are interested.

I think I may have talked myself into trying it in one of my less competitive leagues this year…

Derek Ambrosino
14 years ago

I started writing out responses and realized I have long responses for several of you. For the sake of readability, I will post each as their own post, but some of the ideas mentioned relate to each other.

Derek,

Thanks for posting on this topic as well. I think it is an interesting discussion, and clearly we have several highly intelligent people contributing thoughts and theories. That is what THT Fantasy is about!

I appreciate and agree with your comments (obviously) and think the endowment effect is certainly relevant. A certain quip about stuff vs. um, excrement, by the greatest comedian of all time comes to mind…

Allow me to indulge in a fantasy basketball anecdote or two to offer some additional insight on some of the points made here. A few years ago one of my leaguemates lost our league by a very small margin because he didn’t trade Allen Iverson despite having a huge lead in points. Rounding out a few other cats would have put the league out of reach, but others recognized his situation and lowballed him so egregiously the owner said screw it and decided he had a good enough chance to win anyway and he would not give Iverson away for marginal category security. (More on this idea later) It turned out to be pride turned hubris.

Also, regarding the endowment theory, you know how football translates to television well, but baseball less so. Fantasy players are like that as well. I say this is most dynamic in basketball, but is at play in all sports. There are players you kind of have to own, meaning you watch them produce everyday, to fully appreciate. These are players who do produce all around (at well beyond marginal levels) but don’t wow you. Perhaps the best example of this is the post-Philly Abreu. Torri Hunter was kind of like this too – outproduced his draft pick every year, but you couldn’t get his true value back in a trade.

As a side note, I did not play fantasy baseball back in the 80’s but I’ve always wondered about Coleman in particular. Think about 1986 for a second. Dude hit .239 over 600 ABS, didn’t hit a single homer and drove in 29 runs. But he was 5th in the league in runs and stole 107 freaking bases! How high is that skill set drafted? I guess Henderson, Schmidt, Mattingly, Boggs, Raines, Gwynn, go before him. Maybe Rice, Carter, Barfield too? There are other guys who would go ahead of him as well who would not have panned out, like Brett, Murray, and maybe Gooden. Then, guys who produced amazing value who would not have been drafted so high, like Clemens, Puckett, and Davis. I know the value of steals and power was different relative to one another back then, but it seems that would have been an very extreme case of a decision between very valuable players of extremely different skill sets.

Sorry for the digression…

Derek Ambrosino
14 years ago

Andrew,

Good points. It just goes to show you that you can come at these discussions from many different directions. The observation about being the endowment effect implying it is better to upgrade your starters, consolidate stars into super-duper studs and fill in with undervalued lower tier guys is particularly salient, IMO. The whole goal is to have a plan right? Do I intend to try to maximize the trade value of my bench, or maximize my own replacement value? Two different goals, two different plans.

I will say this (going back to basketball again). The Iverson owner won our league last year – you know how? He drafted Kobe and then traded his second and third round picks for Lebron. He drafts well, so he had competent replacements (There is never really a two-for one in fantasy sports because the second player is really just the marginal value of that player vs. the one you must cut to make room for him – and then the marginal value the guy that second player replaces in your line-up, forcing your weakest starter to you bench. …This may deserve it’s own post because many people don’t seem to understand this). Anyway, the duo of Lebron and Kobe was just unstoppable. He did it again this year, drafting Granger and trading his 2nd and 4th round picks for Wade. Granger just went down for a few weeks, but we’ll see if it works again.

Derek Ambrosino
14 years ago

Eric (if you are still reading),

I have to get drunk before a draft because I’ve really thought about this idea too (I talked in the columns referenced above about artificially manipulating the market for low-volume categories), but I don’t think I’d have the guts to try it unless I was drunk. I’d like to try it just to see what happens… for research purposes. The thing is, if you’re playing in a really competent league, the other owners maybe just effectively collude to stonewall you (kinda like in the Iverson-owner example I mention above) and you’ll be stuck with excess value. But, the allure of the strategy, if executed well is hard to resist. Knowing that you hold single players who could affect the standings so would allow you to get huge value. You let those players out into the market (and control where they go) one by one, thereby depreciating the value of each previous guy you let loose. You might make some huge returns before others figure out what you are trying to do and adapt.

In a muted way, I do this in one of my less-competitive leagues. I think starting pitching is overvalued in that league and relievers overlooked. I can usually beat teams to the wire too. I draft closers at what I perceive to be value and wait on starting pitching. Then I scoop up more closer-converts than anybody. I sell these guys off as the season progresses and my categorical needs become more apparent (sometimes it winds up being starting pitching, last year it was speed). It’s not as extreme as your proposal, but it’s the same principle. I’ve won this league three out of the last four years, and if people would just stop buying closers from me, period, they might hurt their own chances, but in terms of the “league vs. Derek” dynamic, they’d improve the chances of the field.

Side question: What qualifies as collusion via the message board? Is there such a thing? Do any of you have rules against this in your leagues.

For example, if I decided to implement this strategy, and then one of the other owners posted on the message board,“You know what derek’s doing, right?” and outed my strategy, telling others to stonewall me and leave me stuck with my excess SBs. How would you react to that post as a commissioner?

Andrew
14 years ago

Haha, Derek (Ambrosino), that line about having to be drunk to employ that strategy cracked me up.

On an unrelated note, Derek Carty, I noticed Mock Draft Central held its first Experts Draft tonight. Any chance the current NL LABR champ will be participating in one of those drafts in the near future?

Jimbo
14 years ago

I’ve come to really dislike one-trick players. It is SO easy for an “average” 4 or 5 category player to produce better stats.

Who ranked higher on the 2009 ESPN Player Rater, Adrian Gonzalez or Denard Span? Interesting.

Later in a draft I’ll take a sleeper or two that might get onto my team for power or speed, but I still like them to have some shot at a decent average…otherwise they’re a net “no-win” pony. Last year it was A. Hill, this year I like Alcides.

But before taking that sort of player, I ask myself if I’m missing anyone with sneaky value. Abreu is a great example. Asdrubal Cabrera is the poster child for ‘09.

Bresh
14 years ago

I have found lately that after the 1st round, I draft for weaker hitting positions (usually SS, 2B, C).  This affords me the opportunity to obtain elite position players at a premium, without jeopardizing the farm in any one catagory, as usually these players have good R/SB or HR/RBI numbers anyway.  The difficulty I find is targeting the round where I start drafting starting pitching and relief pitching.  I usually wait for a run on relievers, and either go pitcher/player or player/pitcher depending on when I grab my stud starter.  I scour the waiver wire constantly, and am not afraid to take chances on bench players at all, as they usually do get in unless there is an injury anyway.  I won two leagues with this strategy last year, and one the year before.  Give it a shot in your lessor leagues, and let me know what you think.

Derek Ambrosino
14 years ago

Jimbo,

A recurring question underlying much of this discussion is whether one is going to maximize replacement value relative to one’s own team with late round draft picks or attempt to stock potential trade chips. Going back to the example I set forth in a previous article, Melky Cabrera may provide me with better replacement value relative to my own team than, say, Elvis Andrus, especially not knowing the make-up of the roster position-eligibility-wise. But, there’s probably a better trade market for Andrus than Melky because there are lots of players on the wire similar in skill set to Melky “average 3 or 4 category contributors.” But, it’s unlikely there’s much of anything out there in the way of being in the 85th percentile of any single category.