A Few More Stats

Last week I addressed the question, “how much of 2009’s production did each team lose to YE09 bump-ups?” I looked at Mens’ 4.0 and 4.5, and only for Citywide Playoffs matches. This week the analysis is extended to all regular-season matches. Also, since it’s time I took a look at some of my new 4.0 competition, I’ve analyzed 3.5 as well.

In 4.5, the numbers are fairly uninteresting, since there were so few bumps to 5.0. But here’s what the data look like. I have sorted in order from those who return the least 2009 production (on an absolute basis) to those who return the most. Other sort orders may be more appropriate depending on the insight being sought.


Teams shaded in yellow are those that made it to Citywide. As with last week’s analysis, if a doubles line was won by a pair consisting of one returner and one non-returner, the returner was given 40% of the credit for the victory.

In 4.5, it looks like most teams did not lose too much production, as expected. Some stats on the impact of self-rates are shown as lagniappe. It doesn’t appear that self-rates had much of an impact on 4.5 (overall only 13% of matches were won by self-rates), if there is a sandbagging problem you’d expect it to be at the lower ratings.

In 4.0 the numbers are shown below:

Results are broadly consistent with those from the Citywide Playoffs analysis – the top teams have a lot of rebuilding to do, and from about Bear Creek on down the list there appear to be several teams that are relatively well-placed entering 2010.

The real eye-popping results are in 3.5. First let’s look at Citywide Playoffs results:
Two numbers stand out for me – only 10% of lines were won by those who will return to 3.5 in 2010, and 37% were won by self-rates. Clearly, if there is a problem with people underrating themselves, it is in 3.5.

For the full regular season, 3.5 looks like this:

A higher % of lines were won by returnees in the regular season vs. the playoffs, but then you’d expect it since it’s the more-successful teams (i.e. those reaching the playoffs) that suffer most from bumpage. The numbers are MUCH lower in 3.5 than 4.0. I also found it very interesting to see that again 38% of lines won were won by self-rates. I’m sure some of that is legitimate but it just feels high. It would be interesting to see how the won-lost records of the self-rates compare with those of the others…perhaps the subject of a future post.

Those are the numbers, I’d be interested in any other insights that readers can glean from these.