Beatpaths: What’s This About?

We’re getting some increased traffic (clicking my ads will help btw!), so I thought I’d take the opportunity to explain a couple of things regarding what this site is about. I’ve written about it elsewhere, but more’s always better!

A beatpath graph is a graphical representation of all of the games/matches in a season. It pays attention to one stat and one stat only: wins and losses. It doesn’t do any math or assign any equations.

We then generate power rankings. We started with one rule – if a team has beaten another team, it should be ranked ahead of that team. As wins and losses increase, there are less and less possible ways to rank the teams.

It also becomes impossible after only a few rounds, because you might have A beating B beating C beating A. Or more simply, two teams might play each other twice and split the series.

It’s simple: we call those beatloops. We take them out, and we graph what’s left. Since most leagues are highly interrleated, we assume that there’s enough other information in the graph to make up for losing those ambiguous relationships.

So what’s left? A picture of how generally good each team is, by examining wins, losses, and who beat who. It gives more accurate information than records and standings, because they ignore who beats who. I believe it gives more accurate information about playoff seedings, because, at least in the case of the NFL, win/loss records are affected by schedules, and schedules are designed by relying on the strength of teams from the previous season, rather than the current season. This means that teams can be unfairly penalized by having a schedule that is unexpectedly strong, and vice versa. We saw that in 2006 – Dallas should have been in the playoffs, but their schedule was murderous – they didn’t make it.

So what’s the point? If you’ve been using other stats to try and get an understanding of how good a team is, it’s worth checking out the beatpath graph as a supplement. The graph pays attention to wins, losses, who beats who, and nothing else. Same rules for every team. If you’re surprised by what you see, then that means you’re getting valuable perspective.

Plus, it’s just kind of fun to watch how the graphs change from round to round as teams wins and lose. 🙂

4 Responses to Beatpaths: What’s This About?

  1. SB says:

    I’m attempting to do this by hand for my 18-team Fantasy Football League. It may or may not be impossible.

    On the bright side, I’m 5 – 0.

  2. CB says:

    NFL schedules are NOT based on the previous year. They are planned several years in advance, the teams that will be played not the dates the games will be played. Each year a team will play the other teams in its division twice (home and away) for 6 games. The team will play all four teams from one NFC and one AFC conference (two home games and two away for each division) for 8 games. The remaining two games (one home and one away) are played against teams from the same conference.

    As for your example of the Cowboys not making the playoffs because of their schedule. You weren’t specific enough when you said in 2006. This could mean 2006-07 season, in which case Dallas did make the playoffs by winning the division. It could also mean the 2005-06 season playoffs which took place in January 2006 and that year no 9-7 team made the playoffs.

    During the 2005-06 season, Dallas, New York, Washington, and Philadelphia all played each other twice, all four teams from the NFC West and all four teams from the AFC West. So out of a 16 game season 14 of the games were the same. The remaining 2 games (1 NFC North at Home and 1 NFC South away) for each of the 4 teams were as follows:
    Cowboys: at Carolina (w), Detroit (w)
    Giants: at New Orleans (w), Minnesota (L)
    Eagles: at Atlanta (L), Green Bay (w)
    Redskins: at Tampa Bay (L), Chicago (w)

    During the 2005-06 season year the Redskins, by far, had the hardest schedule. Tampa Bay and Chicago both won their divisions.

    During the 2006-07 season, Dallas, New York, Washington, and Philadelphia all played each other twice, all four teams from the NFC South and all four teams from the AFC South. So out of a 16 game season 14 of the games were the same. The remaining 2 games (1 NFC North at Home and 1 NFC West away) for each of the 4 teams were as follows:
    Cowboys: at Arizona (w), Detroit (L)
    Giants: at Seattle (L), Chicago (L)
    Eagles: at San Francisco (w), Green Bay (w)
    Redskins: at Minnesota (L), St. Louis (L)

    Without questions the Giants would have had the tougher schedule. At the time Dallas lost to Detroit, Detroit was 2-13.

  3. ThunderThumbs says:

    #1’s are matched against #1’s, #2’s are matched against #2’s, etc. That’s what I mean by being team schedules being based off of the strength of the previous year.

  4. ThunderThumbs says:

    So yes, it’s true that I generalized about strength of schedule. I deleted some unnecessary rudeness but the rest of it is a good point. To clarify, I’m referring to the 2005-2006 Cowboys. It was their victory over Carolina that made their placement in the graph strong enough to be ranked within the top six NFC teams that year.

Leave a Reply

Your email address will not be published. Required fields are marked *