In this initial pass, we have elected to keep things simple by rating teams as either "good" (.500 or better record) or "bad" (below .500 on the season). With this setup there are eight combinations of three-game-schedules for us to analyze. Depending on what we discover here, we may expand the team rating scheme further on.

Now the other decision we made was to gauage the record of the last game opponent going into that game whereas the this game opponent and next game opponent are rated based on the current week records. This made logical sense to us in that the previous opponent is judged for motivational purposes before the outcome is known.

Finally we set the data range on the queries to only look at games in weeks 9 to 16 (eliminating the problematic last week games) in the theory that this would be prime letdown territory. We'll review the first half of the season games in another article.

The lesson here is home underdogs have been dominant in the middle of a three game swing against good teams. The 45-22 (67%) is strong enough to be worth a run through the 2001/2002 years. Really bad teams (under .350 on the season ) have been a solid 22-8 as home teams overall (mainly of course home dogs).

Home dogs once again are solid, but interestingly away underdogs in this spot have been poor bets against the line, especially if you throw out the bottom category of team (under .350) at which point you have a 26-42 history (38%). Another area of interest is the breakouts by spread range with small underdogs going 36-22, but mid-range dogs going 20-38. Hmmm, we'll watch this scheduling setup too in the retest.

This is the one most commonly talked about, the easy opponent sandwiched between two good opponents. However, there's not a lot here other than the unexpected: underdogs are sketchy (18-31), not favorites.

Small favorites have struggled (18-35), decent to poor teams (under .550) have been excellent as underdogs 33-18. All small samples, all of questionable predictive value going forward, let's see whether the retest delivers a verdict.

Coming off a lesser opponent with two strong ones next up has been a good thing in terms of covering the spread -- 57% all in all, mainly through a 62% record of away underdogs. Worth a second look with the fresh data.

Now we're getting closer to the core of the sandwich game theories, as top teams (.650+) as favorites are just 6-19 (24%) against the line. Of course it's the reverse scheduling scenario (bad-good-bad) than the one we were expecting (good-bad-good). In addition, home underdogs got roughed up playing a good team between two soft ones.

The second below average opponent in a row, with a tougher one in the wings is a strange set of results. Solid teams (.550 and up) are 21-12 as home favorites, but 9-23 as away favorites. Lesser teams (below .550) are 8-21 as home favorites, and 1-3 as home underdogs, but 19-12 getting points ont he road. These kind of patterns have us leaning to the opinion it's all noise, but let's see what the 2001/2002 data has to add.

In the middle of a three game run against under .500 teams, nothing much stands out from the table above, making this the lowest potential scheduling setup of the eight reviewed.

Well, that's the initial read, we'll update this area soon with the 2001 and 2002 breakouts to see if any of the trends that caught our eye held true in the succeeding years.