Monday, August 23 at 1:31 PM ET
Football is almost here, which means quite a bit for the site, including the onset of fantasy football draft season and far more consistent blog entries to help you make better sense of the sports wagering and fantasy sports worlds. During the season, as situations deem necessary or logical, I will live blog through Saturdays and Sundays (or Thursdays, Mondays, etc.). I will also chime in with pertinent topics relative to recent games. And, I can answer your questions. To submit questions at any time, please use the Help link found on any page.
Today's topic stems from a fantasy football question posed through a Help Contact in which one of our users asked about managing bye weeks during drafts. There seem to be three widely-discussed strategies: 1) Never have more than one starter on bye at a time, 2) Line up as many starters' bye weeks as possible to "tank" that week and win others or 3) It doesn't matter.
I believe in #3. Here is part of my response.
Like just about everything mathematical, everything is relative. And, in this case, very little related to intricate bye week strategy matters because it is so hard to execute.
Tanking a game to win all of the others makes a pretty bold assumption that the owner is really good at drafting starters and really bad at drafting back-ups. Very few teams play more than half their games with the starting lineup they originally drafted. And very few teams also hold onto all of their players from the draft – even until the first bye week.
This approach also assumes that every team is average and that your starters would be favored against average teams and your back-ups wouldn't be (though I get that a team may not have enough "back-ups" to field a team). There is not a better assumption to make, but we know that not all opponents are created equally… and what if someone else is trying the same strategy.
And most importantly, there will be some inherent loss in value of the player by trying to implement the tanking 1 game strategy because it requires the owner to draft players in the mid-to-late rounds that match top players' bye weeks. It is highly unlikely that those will also be the top players available with respect to expected points. In fact, since these players would have to come from the same four teams or so, it becomes even more likely that this strategy is reliant week-to-week on one team – meaning that there be other weeks that are “tanked” inadvertently when a fantasy team of Aaron Rodgers, Ryan Grant, Donald Driver, Greg Jennings, JerMichael Finley and Mason Crosby has to play the Jets.
Similar things can be said about trying to scatter bye weeks. It is just so impractical to implement either strategy in a draft – and also comes with value loss.
When I draft a fantasy team, especially with my starters (and top RB and WR back-ups), I completely ignore bye weeks. Value trumps everything. As of the end of the draft, the team with the highest expected fantasy point total on the season should be the team that is most likely to have the best record at the end of the season. It’s very difficult to make any other claim. And it’s virtually impossible to intentionally execute a strategy in real time that could allow an inferior team to have a better expected record.
The only similar argument that I could see with validity is with playoff weeks, not bye weeks. Since there is a premium on the last few weeks of the season in most leagues (something I abhor for the same reason it exists – because it levels the playing field between those that dominate for a full season and those who can luck out late), I can see diminishing the value for players like Peyton Manning whose teams have shown a propensity to sit starters late in the year. There are so few of these instances though, and so much opportunity during the season to find an adequate replacement for those weeks through trades or waiver wire pick-ups that focusing too much on that is probably not "worth it" in a draft.
Again to submit questions to Paul Bessire at any time, please use the Help link found on every page.