Comments on: Mailbag: Last-Place Teams Who Beat the League’s Best

By: MikeN

MikeN — Fri, 08 Apr 2011 20:49:06 +0000

Bulls are next.

The Cavs are underperforming. They have Jamison and Williams.

By: JeremyD

JeremyD — Fri, 08 Apr 2011 19:40:47 +0000

You'd lose a good chunk of your results, but I think the gist of the question would be better phrased by 'How often has the worst team (by however you define worst) beaten a team in the top 1/5th of the league?' I think it is amusing though that from 56-58, any team that the last place team beat would have been in 'the top 6'.

By: P Middy

P Middy — Fri, 08 Apr 2011 13:45:23 +0000

to be fair, that was before Carmelo learned how to play defense for one game.

By: steven

steven — Fri, 08 Apr 2011 01:28:20 +0000

They also beat the Knicks 3 times.

By: Tay

Tay — Thu, 07 Apr 2011 15:22:54 +0000

All you need is the heart of a Champion!!! You can defeat anybody.
http://soundcloud.com/reggiev/champion

By: Mike Goodman

Mike Goodman — Thu, 07 Apr 2011 12:51:15 +0000

This study of course begs the question: What teams finished with the best record and managed to lose to several of the worst teams?

I nominate the 1985-86 Celtics, that dominant Walton-backed team which cruised to a 67-15 record. Among their 15 losses, they were beaten by the 5 worst teams in the East, and twice to NJ.

Their 15 losses came to teams averaging 38.6 wins. Avg SRS of -1.0 .

Now these studies based on W-L records are kind of dodgy, in that a terrible team that manages to beat some great teams may improve their record to where they aren't the 'worst' team that year.

What about a really big data dump that simply compares all team/opponent SRS and finds the least predictable results in a team's schedule? That is, the greatest average difference between expected and actual results. This would obviate the discrepancies of league size, W% vs SRS, etc.

Maybe middling teams would have the most unpredictable results. More likely, those with great home/away disparity. I still go with the '86 Celts.

By: yariv

yariv — Thu, 07 Apr 2011 07:23:21 +0000

Imadogg, you might find it less crazy if you say that the worst record of all time got one third of their wins against the top third of teams...

By: BSK

BSK — Wed, 06 Apr 2011 23:51:45 +0000

To Ed's point, are the Win% calculated at the time of the game or season-to-date (which would be season-end for all but this year)? That seems to matter, since these numbers could jostle. A team may have beat an elite team mid-year only to have that team fall from their lofty perch. Naturally, that would make us cast doubt of considering them to be of elite status, but if their fall from grace was a function of injuries, it might muddy the waters a bit. Of course, any wins in the early part of the year would be equally invalid. I guess the point is that the only appropriate way to do this would be to look at end-of-year Win%, to which the Cavs are yet to necessarily achieve this.

By: Ed

Ed — Wed, 06 Apr 2011 20:44:32 +0000

I appreciate Neil taking the time to research this but the Cavs only have a 1/2 game "lead" on the T-Wolves for the worst record. So they may not end up as the worst team in the league. Which would render this whole exercise pointless.

By: Imadogg

Imadogg — Wed, 06 Apr 2011 20:39:55 +0000

Thanks for the research Neil... wow, the worst record of all time had 33% of it's wins against 3 of the top 6 teams haha. Now that's crazy.