Wednesday, April 14, 2010
Punt your way to the front?Posted by Derek Ambrosino at 4:52am
First, let me proactively respond to what I assume many of you will say to yourselves while reading this column; yes, I’m aware it probably would have been a better idea to write this column during the preseason…
I had some discussion over the past few days about how player value might be affected when by toggling a league’s format between rotisserie (roto) and head-to-head (H2H). The initial question was whether an individual category could be inherently more valuable in a H2H league than in a roto league. Not being aware of any research on this topic, I chose to offer a response that was more strategic than data-driven, which I will now tweak and appropriate into a formal column.
Simply stated, my thesis statement regarding value disparity between roto and H2H leagues is that discounting speed and/or saves can serve as viable strategy in a H2H league but not in a roto league. Secondly, streakier players may be preferable in H2H leagues, but only theoretically.
Perhaps most fundamental to my opinion on player value in respect to H2H leagues is the notion that punting a category is a legitimate strategic direction in the H2H format, so long as such a direction can be accomplished without massive ripple effects. What makes punting possible in H2H formats is the fact that the margin by which you lose a category does not matter and that you don’t actually need to “dominate” your league to win. Let me just offer a little more detail on both these points.
In neither roto nor H2H leagues does margin of victory doesn’t matter. But in roto leagues, margin of defeat is critical because you are still competing against the rest of the field for second through Xth place. H2H leagues are broken into scoring periods which produce binary results—either you get one point or no points.
One can easily win the regular season of a H2H league by winning two-thirds of the possible points. Roto leagues are traditionally less closely bunched than that; 80 points is not likely to be enough to win a 12-team, 5x5 roto league. In fact, I just looked back at my two most competitive 5x5 roto leagues going back to 2004 and calculated that the average league champion took 83.5% percent of the highest possible point total (which would be 100 points in a 12-team, 10-category league). The bottom line here is that there is a margin for error in a H2H league that does not exist in a roto league.
Taking the thumbnail above for the sake of illustration, this means you can forfeit only 16.5 percent of your highest possible score if you want to win a roto league. Taking a “1” in a single category in a 12-team roto league eats up roughly half of that leeway right off the bat. And, mind you, this margin for error could be smaller in any given season.
Now that I’ve beaten these not so revolutionary insights into the ground, we reach the question of which categories would make the most sense to punt, should one choose to do so. To me, the most likely categories are stolen bases and saves—the specialty categories. It’s obviously a bad idea to punt power because a homer is literally a run and at least one RBI as well and therefore by doing so you really take a hit in other categories. I mean, do we even meaningfully differentiate home runs and RBI as distinct categories during the mental process of making draft day decisions? And, while there are some players for whom prodigious power comes at the expense of dead batting average weight, there are plenty of neutral and even plus-batting average options who are plus sources of power as well.
Stolen bases are, by far, the most specialty category on the default offensive menu. So, by punting them entirely you can certainly supplement your power supply and go two-for-one, building strength in two categories as opposed to one. And, while one might be tempted to think that SBs are highly correlated with elite run totals, only one of the seven players who stole 40 or more bases last season scored 100 or more runs (Chone Figgins), while three of the five who hit 40 or more homers scored 100 or more times.
When it comes to pitching, a strategy I’ve implemented in the past, though rarely so, is punting saves completely. It is very difficult to consistently win the pitching rate stats without elite relievers; however, you can gobble up elite setup men for pennies on the closer dollar to fill that role for you. As a bonus, many of them also earn more vulture wins than closers do.
Without having to spend on closers, you can either invest more heavily in your starting staff (reducing the need for the elite relievers) or you can strengthen your offense. Then in the last few rounds of your draft, snatch up the Joba Chamberlains and Daniel Bards, and you’ll be fine. Middle relievers are fickle from year to year (there are many reasons for this, a column perhaps better suited for the non-fantasy side of this site), so there will almost always be options that emerge from nowhere for the plucking.
Another factor making both of these categories punt-able is that one’s advantage in such a category is relatively less reliable from week to week because of the combination of small supply and small sample size. Categories like saves and steals are likely to be close every week by nature. Additionally, a greater proportion of a team’s total prowess in saves/steals is wrapped in a lesser number of players than any other category—another factor exacerbating the potential evils of random variation.
I could beat these two horses even further, but I’ll trust I’ve made my case sufficiently.
The one additional point that I will make though is that H2H leagues also offer more opportunity to make strategic shifts with agility. In roto leagues, you can build cushions in categories and then deal from strength to retool with minimal impact on the overall standings, but H2H leagues are structured such that categorical leads have no carryover value from scoring period to scoring period. Therefore, if you choose to punt steals, you can immediately switch course and retool via trade and free agency if you feel it’s prudent to do so. You will not be facing a pre-existing categorical deficit you must make up before you begin gaining points. You may also wind up winning a week here and there anyway by accident. In essence, you are never truly punting anything for the entire season because the hole you dig is refilled periodically.
Aside from statistical profiles of players, the other thing to consider in terms of how a player’s value may be affected from format to format is whether he is a consistent or streaky producer.
In the theoretical, you'd expect that streakier players are better assets in H2H leagues because their hot streaks have more leverage on that week's outcomes than the consistent but unspectacular contributor. The issues around this are manifold though. First, is consistency actually a discrete, predictable quality? Second, how stark would such a quality have to be to have meaningful impact? Third, the increased streakiness also increases the likelihood of the player producing meaningless surplus value in a given scoring period (not a problem in a roto league, as we just covered).
All in all, I’d consider the streaky versus consistent issue to be way more noise than signal and could not see an instance in which such a quality (real or perceived) could be a deciding factor in how I construct my team. Sure, it may be worth trying to ride a hot free agent for a stretch, or similarly mix and match your roster rotation from bench to starter, but these strategies don’t seem to be any more advantageous or dangerous in H2H leagues than roto leagues. The nature of the H2H league does leave you in more agony over each start/bench decision though.
Thus far, I’ve been fairly mitigating in the way I’ve talked about the possibility of punting, so I guess I should make it clear whether I would actually advise these strategies as opposed to merely acknowledging them as having some undefined amount of philosophical and strategic merit.
My preference is to attempt to stay competitive in stolen bases, but in H2H leagues I am less likely to draft the high-priced speedster than I am in roto leagues (something I’m already fairly unlikely to do). I would draft a Jacoby Ellsbury or Michael Bourn only if I think that player is legitimately the most valuable player on the board … and by a fair margin. But I can’t say that I punt. I still try to build a well-rounded team and spread my stolen base risk (and reward) across a wider cross-section of my roster.
As for punting saves, this is certainly something I consider a very legitimate strategy. I don’t always, or even frequently, practice it. However, that’s actually more because I’ve fallen into a pattern of doing something of the reverse. I often wind up building my pitching staff somewhat backwards, focusing heavily on offense and dominant closers, while picking upside pitchers and streaming opportunistically. I’ll use middle relievers too, if I have the room and they are actually that good, but I often depend on double-digit innings of high-quality bullpen innings on a weekly basis. I also think closers are more of a numbers game than basestealers, figuring that there are only 30 jobs to go around, while every player is a potential basestealer once on base. So, if I have more closers than my “fair share” and my closers are generally of good quality, I think I can win that category fairly consistently, while strengthening my staff’s rate stats.
Derek Ambrosino aspires to one day, like Dan Quisenberry, find a delivery in his flaw, you can send him questions, comments, or suggestions at digglahhh AT yahoo DOT com.