Monday, March 3, 2014

#SSAC14 Betting Panel: Patrick Donovan aka The Sports Boss

As you may have heard, the 2014 MIT Sloan Sports Analytics Conference was in Boston this past weekend, and one of the panels focused on sports betting. Throughout the week, I'm going to be taking a look at what each of the four panelists had to say, including transcripts, related tweets, my own commentary, and (if they're eventually free) YouTube clips.

First up is Patrick Donovan, aka The Sports Boss (Twitter, picks site, blog). Donovan is a tout who appears to work in finance. He is not particularly well known, and may have only been invited onto the panel (at the last minute, it seems) due to being buddies with Chad Millman and the ESPN Insider crew. 

Anything after a bolded name is directly transcribed from the panel itself, anything in italics is me.


Jeff Ma (moderator, table games enthusiast): So, quick question. If you're a handicapper...and you're a good handicapper, I assume, right? You think you're a good handicapper?

Patrick Donovan aka The Sports Boss: I think I'm alright, I'm alright.

Ma: Why don't you just bet your own games? Why would you sell your games to start with?

Donovan: Well, I do bet my own games. But I don't see, there's really no, it doesn't impact me by selling my advice to somebody else if they could profit off or it. So I don't see any reason why not to if I can consistently win and feel comfortable doing that philosophically.

Ma: So do you always bet your games first, and then send your picks out?

Donovan: Yeah, or vice versa. Either way.

Ma: Vice your picks ever move lines?

Donovan: Um, no, I wouldn't say so.

Donovan's website suggests that he "is the BOSS in the sports betting world!"
Ma: Are you, Jay, are you aware of his work at all?

Jay Kornegay (Las Vegas Hotel and Casino SuperBook VP, #makingshitup enthusiast): No.

[laughter throughout the audience]

(video, subscription required)


Donovan: I don't really have a power rating formula for college, for football now, because I think it's very complex, with the strength of schedule. When I was sitting in on the college football panel yesterday, you know they were discussing different ways to calculate a strength of schedule. How do you really calculate it? How do you know, if they say it's the 32nd toughest schedule, what is that based on? Teams' records? Well, who did they play? It's just too tough, I think, to come up with.

Donovan's CFB picks record on his site goes back to 2008.

(video, subscription required)


Donovan: I also look at the probability, so if a team wins like 10 in a row, and especially if some of them were close games, down to the last second, recently, then I like to fade that team. So I've been fading Florida a lot of late. They just keep barely scraping by in these low-scoring games. Very close, and they haven't been covering a lot lately.

Donovan: Virginia, which we talked about in the green room. Today, my biggest play is on Syracuse, because Virginia has won about 11 games in a row; they're statistically...defensively they play well, but offensively they struggle.

Donovan: So I think those are the areas where you have to try to find an edge, whether it's probability, the likelihood of something happening, and the emotional level that a team is going to play with [...]

(video, subscription required)

A few things:
  • Coming into Saturday's game, UVA's offense was averaging ~1.12 points per possession in ACC play, which was second in the 15-team conference.
  • Not expecting Virginia to come out with a high "emotional level" seems reasonable, at least if you ignore the part where a victory would clinch their first outright ACC title in 33 years...
  • ...and also the fact that Saturday was Senior Night in Charlottesville.
  • The Cavaliers had actually won 12 consecutive games coming into Saturday. They have now won 13 consecutive games.

Ma: Patrick, your explanation about Florida and whatnot, I understand it, but it also doesn't give me a warm sense of calm that you know what you're doing.

(The snips in the following section are due to the fact that Chad Millman just would not stop interrupting the discussion with absolute nonsense. This was a regular occurrence throughout the panel. I don't understand how it happened, but he really thinks he's an expert on this all things sports betting at this point, it's pretty unbelievable.)

Donovan: To your point on injuries, I think injuries are one of the most overrated things that handicappers, or oddsmakers, adjust for [...]

Donovan: We can look at a couple of examples here. We could look at Miami, the Heat. LeBron James has missed, I don't know if it's seven or eight games over the last however many years, they're 7-1 in those games. You look at that infamous Spurs-Heat game from a couple years ago where Popovich sent everybody back...Spurs win, in Miami.

Ma: This is not a very big sample size you're talking about here.

Donovan: No, but I've actually researched this. I was discussing with one of the students earlier, where I went back and I looked, in the NBA, at All-Star players. And I looked at when they wind up missing one or two games, and the winning percentage ATS [...]

Donovan: In the NBA, when you're missing an All-Star player, for just one or two games, the rest of the players are able to elevate their game in that small sample size to actually cover, and the number was like 65%, I went back like 10 years. But most of the betting public will go the other way, "Oh, LeBron James isn't playing?" There will be a huge adjustment...

Ma: So do you blindly bet on teams, pretty much, that are missing an All-Star?

Donovan: You honestly could blindly bet on teams that are missing an All-Star in the NBA, and you would be profitable. Blindly bet it. Just follow it out there. Only in the first two games, too.

(video, subscription required)


  1. Hahaha, boss, your best panel yet!

  2. They thus cover High Street as well as on the phone betting along with online betting. Most of the punters prefer betting with this betting site as it is lays much emphasis on its customers. zcode system