Do teams with better road records overachieve in the tourney?

Three members have asked me in the last couple weeks about whether having a good road/neutral record has an impact on tourney performance. I finally got some time tonight to do a quick PASE analysis on the subject. I’m glad Robbie and AC and Rick encouraged me to investigate this.

I only have road/neutral records going back to 2005, but I was able to do the analysis two ways:

  1. I found the median difference between each seed’s overall winning percentage and its road/neutral percentage, then calculated the PASE of teams below the median. My logic is that the teams whose away record is closest to their overall record are the better road warriors.
  2. I took the top six seeds and calculated their PASE against a sliding scale of overall/away winning percentage differences.

Here’s what I discovered with the first analysis:

  • 1 seed (below median .084 overall/away winning rate difference) +.063 PASE
  • 2 seed (below median .114 difference) +.063
  • 3 seed (below median .130 difference) +.406
  • 4 seed (below median .135 difference) +.156
  • 5 seed (below median .150 difference) -.473
  • 6 seed (below median .157 difference) -.313
  • 7 seed (below median .160 difference) +.031
  • 8 seed (below median .165 difference) +.156
  • 9 seed (below median .150 difference) -.063
  • 10 seed (below median .165 difference) +.031
  • 11 seed (below median .148 difference) +.156
  • 12 seed (below median .130 difference) +.063

If you take the teams with away records below the median in terms of their difference to their overall record, they perform just +.004 above seed expectations. But most of the underachievement happens at two seed positions—the fifth and sixth seeds. For them, overachievement means beating threes and fours. So they may be losing at the expense of other good road warriors. If you just focused on the one through four seeds below median overall/away wining percentage difference, those squads overachieve at a +.172 PASE rate. So it’s fair to say that playing well on the road does correlate with overachievement in the tourney, at least for the highest four seeds.

I figured I’d take the analysis one step further and calculate whether the tourney performance of teams improved with corresponding improvements in their away records. I took the top six seeds and did a PASE analysis for six ranges of differences between overall and away records—less than a .150 difference, less than .125, less than .100, less than .075, less than .050 and less than a .025 difference. Here’s what I discovered:

2013_road_value

There were 119 one through six seeds with less than a .150 difference between their overall and road record. They were actually -.039 PASE underperformers. The 94 high seeds with overall/road record differences of less than .125 were slight +.024 PASE overachievers, while those with differences less than .100 were slight -.020 underachievers. Once the overall/road record difference gets less than that, however, the rate of overachievement among high seeds accelerates. The 50 teams with a differential of less than .075 are +.074 PASE overperformers—and the 34 teams with a overall/road difference of less than .050 are sizable +.342 PASE overachievers.

Bottom line: I’m going to keep my eye on the road/neutral records of the higher seeds. They may just help me determine which among them are more likely to beat seed expectations.

This entry was posted in Basic Concepts, Tourney Trends. Bookmark the permalink.

12 Responses to Do teams with better road records overachieve in the tourney?

  1. Adam says:

    Would you be able to figure out what kind of success/failure mid-major teams that are seeded 5 or higher have had in the past however many years? Reason I ask is because teams like New Mexico and Gonzaga (potentially a 2 or 3 seed, potentially a 1 seed this year respectively) have disappointed in the tournament as high seeds multiple times. Would be interesting to see some data on that.

  2. Ryan Tressler says:

    Great stuff Pete, really liked the new champ credential post as well . . . I cannot wait for conference tournaments to start (tomorrow for Big South and, more interestingly, Horizon conferences!) and then Selection Sunday (less than two weeks away!!!)

    Speaking of the Horizon Conference, if Detroit and Valpo make that tourny final, its gonna be a heck of a game, as they are both teams are very good (both capable of possibly getting a win or two in the ncaa tourny if they make it), but probably just outside the bubble

    • ptiernan says:

      I’ve got a feeling both will have to win the Horizon to get in the tourney. I wouldn’t want to face Detroit.

      • Ryan Tressler says:

        I completely agree, which is why I think that game (if we get it) will be a hidden highlight of championship week

  3. ptiernan says:

    Since I’m right outside Detroit, I would have a rooting interest in that game.

  4. Steve says:

    What do you define as a road/neutral game? Is it literaly any game not played at home? So for example, if Kansas plays a game in Kansas City, would you have included that as a road/neutral game?

    • ptiernan says:

      Unfortunately, I don’t differentiate the proximity of a neutral game to a team’s campus. So it’s literally any game not at home.

  5. AC says:

    Thank you! This was interesting to see. I’m assuming you could make the same argument that you made for the top 4 seeds for 10-12 seeds as well.

Leave a Reply

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

* Copy This Password *

* Type Or Paste Password Here *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>