Week 26 Mixed-League FAAB Estimates
The FAAB bids this past week at the NFBC’s Main Event mirrored the regular season, almost dead. Even so, I’m back for another week of estimates for the upcoming bidding cycle.
First off, here is a snapshot of last week’s results.
My forward looking FAAB projections got murdered this week
Name: Proj, Actual
Mejia: $19, $17
Voit: $61, $12
O'Hearn: $10, $7
Alcantara: $13, $7
Wainwright: $25, $6
Shoemaker: $15, $5
Valdez: $24, $4https://t.co/9oSIMQLCQoI need to dive in and do some late season adjustments
— Jeff Zimmerman (@jeffwzimmerman) September 11, 2018
On average, the bids were significantly higher than expected except with some rookie call-ups. I went back to the drawing board and completely reworked my formulas with some historically acceptable results.
My current quandary is making the estimates more usable. Before next season, I can reexamine the information with fresh eyes and possibly explore other options.
The new projection formula lowers many of the estimates into the $1 range. These are acceptable values with just a few dollars left and fewer teams competing.
Here are this week’s estimated FAAB with the ownership percentages from CBSSportsline and bid estimates for NFBC Main Event ($1000 Budget, low bid of $1).
| Name | Previous Ownership% | Current Ownership% | Change% | Cost |
|---|---|---|---|---|
| Wade Miley SP | MIL | 30% | 42% | 12% | $3 |
| Jake Junis SP | KC | 62% | 73% | 11% | $12 |
| Adalberto Mondesi 2B | KC | 33% | 43% | 10% | $2 |
| Brad Keller SP | KC | 35% | 45% | 10% | $2 |
| Luke Voit 1B | NYY | 31% | 40% | 9% | $20 |
| Framber Valdez RP | HOU | 16% | 25% | 9% | $9 |
| Felix Pena RP | LAA | 11% | 20% | 9% | $1 |
| Jose Urena SP | MIA | 23% | 32% | 9% | $1 |
| Carlos Martinez SP | STL | 71% | 79% | 8% | $13 |
| Kevin Kiermaier CF | TB | 19% | 27% | 8% | $9 |
| Greg Holland RP | WAS | 13% | 16% | 3% | $1 |
| Ryan O’Hearn 1B | KC | 12% | 15% | 3% | $1 |
| David Fletcher SS | LAA | 9% | 13% | 4% | $1 |
| Yoshihisa Hirano RP | ARI | 8% | 12% | 4% | $1 |
| Adam Frazier LF | PIT | 8% | 11% | 3% | $1 |
| Jorge Lopez RP | KC | 4% | 10% | 6% | $1 |
| Tommy Hunter RP | PHI | 3% | 6% | 3% | $1 |
| Rowdy Tellez 1B | TOR | 3% | 5% | 2% | $1 |
| Ji-Man Choi 1B | TB | 2% | 10% | 8% | $2 |
| Ty Buttrey RP | LAA | 1% | 3% | 2% | $1 |
The biggest item I’ve noticed examining the list was the low bids on closers (Hirano, Hunter, Holland, and Buttrey). I rechecked my math and the “error” is with the CBS ownership rates. The four saw their ownership rates vary by only 2% to 4%. Earlier in the season, a new closer may see a 50% jump. Just a few teams care right now.
On the same note, the biggest mover was Miley with only a 12% move. Again, that value is extremely low compared to earlier in the season.
Overall, the results will hopefully be better but the low late-season activity is messing with my calculations. I need to see how to adjust the final two weeks of data to help with next season. Until then, happy fishing.
Jeff, one of the authors of the fantasy baseball guide,The Process, writes for RotoGraphs, The Hardball Times, Rotowire, Baseball America, and BaseballHQ. He has been nominated for two SABR Analytics Research Award for Contemporary Analysis and won it in 2013 in tandem with Bill Petti. He has won four FSWA Awards including on for his Mining the News series. He's won Tout Wars three times, LABR twice, and got his first NFBC Main Event win in 2021. Follow him on Twitter @jeffwzimmerman.
A thought to adjust for this lack of late-season activity: instead of just “Change %”, also “% of Total Pickups”. Maybe even “% of Total Pickups Above a Baseline of X” so that a constant amount of necessity-type maintenance moves aren’t counted toward the total.
For instance, this past week, lets say Wade Miley was picked up in 12% of 5,000 leagues, so 600 pickups. If there were 2,000 total pickups, and maybe 250 of those were due to necessity (replacing a garbage C or SS that is injured/benched with another garbage C or SS that is healthy and playing, therefore unrelated to any increased demand for the player that was picked up) then there were 1,750 demand-based pickups. 600/1,750 = 34% of total demand-based pickups. But say, in week 4 if a guy has a Change % of 12%, it’s still 600 pickups, but the percent of total demand-based pickups would change the denominator to something like 7,500 total – 1,500 maintenance=6,000 demand-based pickups, which works out to 600/6,000=10%
Maybe this is looking too much into it, and is too convoluted. You’d have to calculate, or predict, the demand-based pickups each week which might only give diminishing returns, so can be ignored. Either way, still lots of work to do and I love that you’re putting in the effort, and luckily it’s for the exact format I play. So, thanks!