Sunday, August 25, 2013

How to draft your team, not a team.


Before starting any fantasy football analysis, we need to start at the beginning and ask a simple question: What is your goal?  If your goal is to win your league, what does that mean?  It might mean to make the playoffs, then win the playoffs.  It might mean to win the regular season, with playoff success as a bonus.  My assumption is that to win a standard league, you need to make the playoffs then win the playoffs.  That means 8 wins, plus wins in weeks 15 and 16.  That’s the goal: 8 wins, plus two.  Throughout draft prep and the season, think about this goal first.

The most important event of the fantasy season is the draft. That's no secret. You can make shrewd trades or pick up a stud off the waiver wire, but those moves only supplement what you have already rostered during the draft.

There are lots of tools that will help you win a draft. Do enough mocks, and you will become good at mock drafting. But most of us are not in a hundred leagues we care about equally, we have one league we care about more than any other. So the problem is not winning a draft, the problem is winning your draft.

Do enough mocks, and you settle into a pattern. You find the same sleepers every time. And that's fine if your goal is to be better than average over a large number of drafts. But if your goal is to win one particular draft, mocks don't help.

To win one particular draft, you need several tools. Some of these are available on various fantasy sites, some you will have to do yourself. Consider it training camp.

1. A comprehensive draft sheet
I will not call it a "cheat sheet", because typical cheat sheets are woefully inadequate. This needs to be a page with tiered positions, point projections per tier, and tier round ranges.

2. A schedule strength sheet
Think about your goal: win enough weeks to make the playoffs, then win the playoffs. To make the playoffs you need 8 wins. So what you need to know during the draft is what those 8 weeks are, so you can optimize. Listing byes doesn't cut it.

3. A late-round streaming guide for each position.
You don't necessarily know which position you will be short on, if you are value-drafting. So make sure you know how to stream QBs or TEs or your third WR if you need to. Don't wait on RBs, streaming them isn't pretty.

First, the draft sheet.  A typical “cheat sheet” has a list of players at each position, with a bye week next to each name.  Can this be less helpful?  I prefer to vomit into a bucket and use that instead.
Any self-respecting draft sheet has tiers; players in groups of like-valued players.  For instance, Calvin Johnson and AJ Green might make up your WR Tier 1 (or WR1, as I’ll be using), while Brandon Marshall and Dez Bryant are in your Tier 2 (WR2).  This is better, but still inadequate to differentiate ourselves from our drafting opponents.  And importantly, this is as far or farther than nearly every out-of-the-box “cheat sheet” will go.

So we need a next step.  We need to keep in mind our Goal.  Let’s say we are sitting in round 4, and you have rostered three guys.  You have 5 guys left in a tier of WRs to choose from.  Which of those WRs will help your specific roster the most?  This isn’t about covering byes, it is about optimizing your lineup to win 8 weeks.  What additional info do we need to achieve this goal?

Basically, we need to quickly know what our strongest and weakest weeks are as we are drafting.  The info needs to be quick and clear, and it needs to change as players are added to the roster.  And it needs to show how adding certain players would affect it.  In the end, we should have maximized our odds of winning 8 weeks, plus the playoff weeks of course.
The first step is to take a look at the playoff weeks and eliminate some teams.  There’s no way to know how good defenses are going to be by the end of the season, but instead of predicting a top ten we can reasonably predict a couple teams to stay away from.  In the past 4 years, only 2 teams have finished in the top 2 of scoring defense, then outside of the top ten the following season: the 2010 Cowboys and 2011 Packers.  But those teams weren’t supposed to be good in those years; a drop-off was expected for each, though not as drastic as what occurred.  The top two scoring defenses in 2012 were Seattle and San Francisco, and both are expected to be excellent again.  So in the interests of winning in the playoff weeks, we should avoid players on the Giants (Sea Week 15) and Bucs (SF Week 15), and especially the Cardinals (@Sea Week 16) and Falcons (@SF Week 16).  Another options would be to draft those players but ultimately trade them prior to the playoffs, but the point stands: those teams face tough matchups in must-win weeks, so they should not be on our rosters for those weeks.  This is great info to use if you are debating between Roddy White, Larry Fitzgerald, and Dez Bryant (GB Week 15, @Was Week 16).

The first draft pick sets the tone for which weeks will make up your 8 wins.  If you select Adrian Peterson, you should expect to win Weeks 1 and 9 (@Det, @Dal), and he give you potentially a good matchup in the playoffs’ first round (home against Philly).  But his bye is Week 5, and he has tough matchups in weeks 2, 4, 11, 13, and 14, plus the potential super bowl week 16 (@Cincy).  So after drafting Peterson, your priority in subsequent rounds should be to strengthen at least one of the five weeks in which Peterson has tough matchups or a bye, without sacrificing any of the weeks where Peterson has more neutral matchups.  You may value Cam Newton or other Panthers a bit more, because they have favorable matchups in weeks 2, 13, and 14 to cover weaknesses in Peterson’s schedule, and it may help you solidify weeks 4 and 11 as weeks you don’t care about and won’t optimize for.  Panthers and Packers have week 4 byes; Cowboys and Rams have week 11 byes.  All else being equal, you should move players from those teams up in their tiers because they are more valuable to your team.

Knowing what weeks are your strongest and weakest has an added benefit.  Say you lose your first week.  No problem, right?  It’s a long season.  Then you lose your second week.  Is it time to panic?  If you know your weekly strengths and weaknesses, and the first two weeks were your strongest, then yes it is time to panic.  Have a firesale!  But if the first two week were your weakest, then hold on for a while longer.  It’s the fantasy equivalent of losing road games in the NBA playoffs: the season doesn’t start until you lose a strong week or win a weak one.

This is all great, but how do you determine what is a strong week versus a weak week?  This is where some personal investigation is needed, based on league scoring settings and history.  In my league, we have had the same scoring settings for the past three years, so I can easily determine what the average point total is that puts me in the upper half of weekly scoring, meaning I have a greater chance of winning the week than losing it (this all depends on opponent, which is uncontrollable; I’m talking probabilities here).  For simplicity, let’s say that the weekly tipping point is 100 points.  Score 100 every week, and you will score more than at least 5 of your leaguemates in a 10-team league, and possibly less than 4, which amounts to at least a 56% chance of victory, which over the course of the season means at least 7.8 wins.  So 100 points should be your target per week.

And how to get there?  That is where the point projections per tier come in.  Let’s say I attach 10 points to my WR2 tier.  My next step is to define up to ten defenses that I believe will be the best scoring defenses and up to ten that will be the worst.  To take it a step further, I could define them based on position, and have different lists for RBs versus QB/WRs.  Then I look at the NFL schedule, and for each time a player has a tough matchup I subtracts 2 points for that week, and for each time they have a good matchup, I add 2 points.  Using a spreadsheet, I can quickly add the columns from the players I draft (or players I might draft) to see what my totals are for each week.  I can use placeholders for kicker and defense, since of course I’m waiting until the end to draft them, and easily see what my 8 win weeks are, and which ones I need to augment.  Late in the draft, when choosing backups and filling out the roster, I can make educated decisions about which players will help my team the most, not just some generic team.
 
It takes some work, but it personalizes your team like no other method.  In addition, it gives you benchmarks to measure your team during the season.  As defenses reveal themselves to be great (or terrible), you can adjust your projected scores and evaluate trades based on getting you the points to reach 8 wins, and then set yourself up to succeed in the playoffs.  That the goal, after all.