While contemplating the result from an earlier blog, which was that home teams had higher handicap-adjusted margins and won at a rate significantly higher than 50% on line betting - virtually regardless of the start they were giving or receiving - I wondered if the source of this anomaly might be that the bookie gives home teams a slightly better deal in setting line margins.

More specifically, I wondered if home teams that were rated an x% chance of winning as determined by their head-to-head price, would receive a more favourable handicap than away teams priced equivalently. So, for example, since home teams that are rated about 30% chances of winning (ie that are priced at about $3.15 if there's a 5% vig) have on average been given 18.4 points start, would away teams that were similarly rated and priced on average receive, say, only 16.5 points start.

If this were the case for that specific example and more generally, it would go some way to explaining the unexpectedly high success rate of home teams on line betting.

Cometh the hypothesis, followeth the data analysis:

Rarely does an analysis so distinctly demolish an hypothesis. What this table shows is that home teams, rather than receiving a leg up from their handicapping relative to away teams, actually receive ... well whatever parallel construction you want to devise that means the opposite of a leg up. An arm down, perhaps.

For those games where the home team is priced between about $1.30 and $3.45 - so their implicit probability of victory is between about 27.5 and 72.5% - the home team receives fewer or gives more points start than do equivalently priced away teams. The magnitude of the extra handicap borne by home teams varies, but generally lies in the range 0.5 to 1.5 points.

Now home teams are priced between $1.30 and $3.45 in about two-thirds of all games. In the remaining one-third, home teams do enjoy slightly more favourable handicapping treatment than equivalently priced away teams, but the size of the benefit is not sufficient to outweigh the burden home teams carry in the other two-thirds, with the result that, averaged across all games, home teams carry an additional 0.4 point handicap when compared to away teams with an equivalent price.

So now we've a riddle to add to the previous enigma.