Posted by Doug on May 23, 2007
I was watching the NFL Network's minicamp show about the Cardinals last week, and they mentioned that it was 105 degrees as the Cards worked out at their Tempe facility. As a former Tempe resident, I'm well acquainted with 105-degree May afternoons. For some reason, this is not very widely reported, but we Arizona denizens happen to know that the heat there is actually rather dry and thus doesn't quite feel so hot. Still, 105 is 105.
Anyway, it got me to thinking. Back before the lights were installed at Wrigley Field, the Chicago Cubs could generally be counted on to play much worse in the late part of the season than they had in the early part of the same season. I can't put my hands on a cite at the moment, but I recall that the evidence was pretty stout. It is simply a fact that the Cubs consistently played worse at the end of the year in those years. Sports-obsessed fifth graders certainly appreciated all those 1:35 starts in June, July, and August, but it isn't too surprising that they would eventually take a long term toll on the Cubs' players. It's not that they couldn't handle the heat on any given day --- in fact they probably had a slight advantage during those hot day games --- it's that they eventually got physically worn down during the long season and didn't have as much energy left in September.
Obviously, the Arizona Cardinals don't play 81 games per year in the blistering heat, and it's probably irrelevant that they actually play their games in a dome now. The fact is that they do live and work in the heat on a day-to-day basis. I don't know exactly how much they practice outdoors, but September and even October can be very hot in Phoenix, and maybe that takes a toll on the Cardinals.
Look at this remarkable split of all Cardinal games since the introduction of the 16-game schedule in 1978.
Early-season Late-season ================================================== in St. Louis 41- 70 0.371 20- 19 0.512 in Phoenix 83-146 0.362 22- 53 0.293
Late-season is defined to be any game in the last four weeks of the regular season.
That's a pretty striking set of splits, but there are reasons to believe that they don't support my theory as well as it might at first appear. First, the strangest number there isn't the poor late-season winning percentage by the Cardinals since they moved to the desert; it's the (relatively) great winning late-season win percentage they had in St. Louis, which probably doesn't have much to do with the weather.
Also, if the heat were really the culprit, we'd expect to see similar splits for other hot-weather teams. With the notable exception of Dallas and to a lesser extent New Orleans, the other hot-weather teams generally have had late-season records very similar to their early-season records:
Early-season Late-season ================================================== Dallas 197-144 0.578 55- 60 0.478 New Orleans 159-181 0.468 49- 67 0.422 Miami 202-137 0.596 68- 48 0.586 Tampa Bay 139-201 0.409 46- 69 0.400 Atlanta 147-194 0.431 49- 66 0.426 Houston Texans 18- 42 0.300 6- 14 0.300 Houston Oilers* 99-109 0.476 35- 37 0.486 Jacksonville 75- 69 0.521 27- 21 0.562 * - Tennessee stats not included.
One more interesting note: at least according to the difference between early-season and late-season winning percentages since 1978, the team that has the most history of wearing down late in the year is one that plays in another city with a unique characteristic which could potentially explain late-season fades:
Early-season Late-season ================================================== Denver 217-123 0.638 60- 55 0.522
I ran a few checks to see if that split was sensitive to the definition of "late-season" or if it was just that they had played more road games late in the season, or had for some random reason played stronger teams later in the year. But no, it's legit.
That doesn't mean it's not a fluke; it just means that it doesn't seem to be a scheduling fluke. The Broncos really have played worse, all things considered, late in the season than early in the year.
Does anyone know if the Nuggets and/or Rockies have historically shown a similar pattern?