Win Now Trades and Playoff Win Probability

Six months ago, I examined in-season fantasy scoring for dynasty leagues to determine how much a “win-now” trade impacted win probability. I suggest glancing at that article before reading this follow-up, despite the exaggerated title I deeply regret. I hoped owners could use the information to decide whether or not pulling the trigger on a trade that sacrificed future value for present value (or vice versa) was worth it.

This article expands my analysis to playoff scoring as well. After all, owners aren’t making win now trades with just the regular season in mind. I will also expand my study to multiple league settings. This playoff analysis, in conjunction with my last article, should give owners a better idea of the impact a win now trade has on your dynasty team.

Regular Season vs. Playoff Scoring

Playoff scoring samples are much smaller than regular season samples for fantasy leagues, so I expanded my sample size to multiple leagues. I also separately examined leagues with different roster sizes and scoring settings. First, let’s compare playoff scoring to regular season scoring for the same setup used in my previous article, a fairly standard dynasty league letting: 0.5 PPR, QB/RB/RB/WR/WR/WR/TE/FLEX/DST/K:

[am4show have=’g1;’ guest_error=’sub_message’ user_error=’sub_message’ ]

Needless to say, scoring is higher in fantasy playoffs. More interestingly, the average and median margin of victory are larger. This makes sense, since there are a higher percentage of high scoring “boom” weeks from teams that make the playoffs. Lower scoring teams have been filtered out by that point, so the only teams that remain are those most capable of scoring big any given week. In the regular season, in order to turn a loss into a win with a new player acquired from a trade you’d have to close out at least a 20 point difference half the time in 0.5 PPR in the regular season. In the playoffs, the threshold increases from 20 to 24, making it more difficult.

Though there were more frequent high scoring weeks in the playoffs, those extreme scores were slightly closer to the mean, since the average playoff score is higher than the average regular season score. This is an indicator that the distribution of playoff scores has a slightly different shape when compared to the regular season. For example, here’s the playoff scoring distribution for a full point PPR dynasty league with an extra flex:

I created three playoff scoring distributions: The one above, one for the 0.5 PPR league setup described previously, and one for a full point PPR league with the same settings as the 0.5 PPR league otherwise, using multiple leagues in each sample (with help from @BrianMaloneFF and @DynastyKester). Even though the average scores varied across all three league setups, it turns out that the playoff scoring distributions were roughly the same shape!

I’m hesitant to use that result to draw sweeping conclusions based on only three league settings tested, since I would imagine the standard deviation of leagues with exceptionally large starting lineups or high scoring settings would be significantly larger. Unless you’re in a league like that, I’m willing to conclude that while the exact win probabilities may vary slightly, you can expect the following results to apply to your league.

Win Probabilities

Let’s examine the difference in regular season and playoff win probability tables for the 0.5 PPR league setup referenced earlier:

A refresher on how to read these tables: During the playoffs, scoring 120 points in a week gives you a win probability of 55%. Increasing your point total by five PPG would increase your win probability by 9%. (I rounded the percentages this time for easier reading.)

Observations

Win probability is lower in this point range during the playoffs, as would be expected. For low point totals, the win probability increase for increasing expected PPG is less during the playoffs than during the regular season. If you’re scoring that low, you’re probably not making the playoffs anyway. If your team lucked into the playoffs despite low weekly point totals, it will be even harder to turn a loss into a win.

The sweet spots (places where win probability increase per point added is at its greatest) are located at slightly higher PPG values in the playoffs. This is because the average score is higher in the playoffs, changing the location of the hump of the data distribution. What’s more interesting, however, is that in the playoffs these sweet spots are slightly sweeter: Aside from non-playoff-caliber PPG outputs, the peak win probability increase for any given expected PPG increase during the playoffs is larger than in the regular season. Remember when I pointed out earlier that the extreme scores were closer to the mean? This is a manifestation of that observation. More weekly scores are closer to the (elevated) mean during the playoffs.

These tables apply only to one specific league setting. What follows is my attempt at a table that allows anyone to apply these results to their dynasty league:

This chart is read just like the ones above, except expected PPG is expressed as a percentage relative to the median playoff score for your league. For example, if you believe the median playoff score is 120 PPG in your league and your team’s median weekly PPG is 126 PPG, then that translates to 5% above the median (126/120=1.05.)

In my last win probability article, I also included a table for those considering being on the other end of a win now trade despite competing for the playoffs, which is something I do often. Here is the playoff version of that table:

Conclusion

After being on the win now end of a win now trade, you will see one of three possible fantasy playoff matchup outcomes:

Outcome one: You win, but you would have won even without the player or players you added.

Outcome two: You win, and the player or players you added were the difference.

Outcome three: You lose, and you would have lost even without the player or players you added.

I hope this analysis helps you decide how likely it is to turn outcome three into outcome two. Compared to the regular season, each point added leads to a bigger win probability increase, during the playoffs, and a win in the playoffs is worth much more than a win in the regular season. It must be noted, however, that the playoffs provide a much smaller window to allow for turning a loss into a win, made even smaller by the fact that a certain number of losses ends your playoff run.

A 10% win probability increase from a trade made prior to week one gives an expected win increase of one over ten weeks, but only a 0.6 expected win increase over the following six week period. Only half of those are playoff weeks in most leagues and that’s assuming you first make the playoffs and then survive long enough in the playoffs to play three meaningful playoff matchups. Still, the value and glory of a playoff win could make a win now trade worth it, especially if you’re in the sweet spot where win probability increase per point added is at its most efficient.

Analyzing playoff win probability made me reconsider my stance in my previous article where I declared the win probability increases to be too low to be worth it, and in fact I advocated for the opposite: Taking advantage of other owners trying to make win now trades by being on the other end. My stance remains largely unchanged, but instead of going out of my way to rid myself of win now assets before the deadline, I’ll more frequently be content riding them out during the playoff run.

I encourage you to take a look at these numbers and make your own decision when considering what the best course of action is for your franchise as we approach the stretch run of the fantasy season.

[/am4show]