Temper Your Expectations: Targeting Boom/Bust Players
In Part 1 of “Temper Your Expectations”, I used the standard deviation of week to week projections for top fantasy football players, and introduced a metric called Points Below Average Projection (PBAP). PBAP gives us a baseline that we can use to recognize when a top fantasy player is expected to have a game that is statistically lower than their normal range of outcomes.
Now that there is a numeric guideline in place to help fantasy owners recognize “temper your expectation” scenarios, we can explore how 4for4's in-depth weekly projections can help us combat a likely down week from a top player by targeting boom/bust candidates at other positions.
Targeting a boom/bust candidate or two is especially effective when an expected drop in production from a top tier player will likely make us an underdog in our head to head match up that week. 4for4's Full Impact Weekly Assistant Coach is the perfect tool to determine if we are, in fact, an underdog or a favorite.
Recall our example of Drew Brees playing on a cold, wet night in Seattle. Let's assume that 4for4 has projected Brees for 17 fantasy points, but his average projection is 20.3. This 3.3 PBAP is well above our 2.19 PBAP threshold and after comparing teams in the Full Impact Primer, we see that we are an underdog. All of our start/sit decisions are fairly cut and dry except for our flex position.
We're trying to decide between two players that are both projected for 8 fantasy points, but we want to start the player that has the best chance to blow up, even if it means that he is more likely to post a dud. Because we're the underdog, the player that's going to consistently score 8 points every week does us no good. By going beyond projected fantasy points, and digging deeper into 4for4's projections, we can discern the boom/bust candidate from the Steady Eddie.
Identifying Boom/Bust Candidates
By definition, a boom/bust candidate is a player with a high ceiling, but a low floor. What will that player look like? Most of the time (all of the time?) it's going to be someone that has a chance to score a touchdown or two on relatively few touches.
An examination of fantasy output and standard deviations within that output from fantasy starters over the last 3 seasons gives us a reference point to how often, or at what rate, each position should be scoring touchdowns and fantasy points. If we can pinpoint players that are projected to score fantasy points or touchdowns at higher rates than usual, then we have successfully quantified what a boom/bust candidate looks like in 4for4's projections.
* Quarterbacks see enough volume per game, have a high enough floor, and can usually be predicted with such accuracy, that the position almost never makes for a true boom/bust play.
Because individual scoring is contingent on so many contributing factors, we can look at fantasy starters as a whole to get an idea of the rate at which we can expect running backs to score fantasy points and touchdowns. Here's a look at expected scoring rates for top-24 running backs over the last three seasons:
Expected Scoring Rates for Top-24 RBs, 2011-13
|Fantasy Points/Touch (FP/t)||Touches/TD (t/TD)|
When we look at weekly projections, we can easily identify boom/bust candidates based on their projected FP/t and t/TD ratios. Any ratio greater than one standard deviation away from the average denotes a player that is likely to touch the ball in scenarios that favor scoring, our classic boom or bust target.
No projection will ever have have a t/TD ratio 2 standard deviations below the mean, but once in a while the projected FP/t will be greater than 2 standard deviations above average. If you can spot a running back with a projected FP/t greater than 0.93, that is the ultimate boom/bust back.
In 2013, 4for4 projected 1291 instances in which backs would have at least 5 touches, and 2.3% of the time a player's projected FP/t was above 0.93. The players with abnormally high FP/t numbers that showed up the most were Joique Bell, Shane Vereen, Darren Sproles, and Danny Woodhead; all of these players were backs that didn't see a ton of touches, but caught a lot of balls, usually in great scoring situations.
If we think about the type of backs that the projections suggest are boom/bust candidates, it makes sense. Running backs that rely on catches out of the backfield for a large portion of their workload are often in space and are good check down options when passing lanes are tight in the red zone. Our projection analysis not only highlights boom/bust candidates from a numbers perspective, but also gives us an idea of the player type that is likely to have huge upside in a given week.
Expected Scoring Rates for Top-36 WRs, 2011-13
|Fantasy Points/Reception (FP/rec)||Receptions/TD (TD Rate)|
Touchdown rate for wide receivers has such high variance that even the most efficient scorers are going to fall within one standard deviation. Projected FP/rec can be used as an indicator of a potential big play receiver, and a good TD rate is inherently going to be accounted for in that ratio.
A good boom/bust wide receiver candidate is going to be projected above 2.41 FP/rec, but an ideal situation is a receiver projected more than 2 standard deviations above the mean, or at least 2.8 FP/rec. Last year 4for4 projected a wide receiver to score more than 2.8 FP/rec just 1.6% of the time, but those rare instances highlighted players like Terrance Williams, Kenny Stills, and Justin Hunter.
Keep in mind that with both running backs and wide receivers, we are interested in the scoring ratios. Most players that we will consider in these boom/bust situations will be projected for a fraction of a touchdown, but because we will be comparing two players with similar median projections, we simply want the player that scores at the highest rate. 4for4's weekly projections provide all the information needed to calculate those ratios.
Looking at scoring rates can give us a numeric representation of who actual boom/bust candidates are in a given week and possibly suggest specific player types that we should target when we need to seek upside over consistency. Uncovering these players requires fantasy owners to scrutinize projections beyond just fantasy points, but understanding projected scoring rates gives us a leg up in a game that requires pushing a bunch of small edges.
When it comes to boom/bust plays, fantasy owners must remember that we are seeking a projection that is actually a statistical anomaly, meaning that huge variance is built into the projection. A boom/bust candidate that doesn't recognize their frequent scoring potential usually won't make up fantasy points with volume. When our player does end up near the floor of their range, take solace in the fact that, as the underdog, it was still the correct decision.
Fantasy football decisions should never come down to rigid rules or restrictions, but a diagnosis of past projections, historical scoring output, and mathematical analysis has helped us formulate some guidelines to recognize the often spoken about “temper your expectations” scenario and how we can possibly withstand a possible down week by identifying and plugging in boom/bust candidates.
For the final chapter of “Temper Your Expectations”, I'll examine actual fantasy scoring of running backs and wide receivers from the last two seasons to see if our projection model accurately predicts players that score at a higher rate than usual.