Monday, August 11, 2014

The Long Con of @RJinVegas: #MATH

RJ Bell is the founder and CEO of, a sports betting website that includes forums, odds, contests, and picks (both of the free and for purchase varieties). It is one of the most successful sites of its kind, in no small part due to the popularity of Bell, who has accumulated over 80K Twitter followers and established himself as the mainstream media's go-to "expert" for stories involving gambling on sports.

An American success story if there every was one. That's only one side of it, though. In some corners of the internet (including this blog), there is a feeling that Bell has not found his success in a legitimate way. This is the seventh post in a series detailing the reasoning behind that particular feeling. If you'd like to start at the beginning, it's here.

Beyond lacking any understanding of how betting markets actually work, RJ Bell is also not very good at simple math. I can prove this in under 140 characters:
This tweet is concerning for a long list of reasons. For our purposes, I've narrowed it down to four.

Concern #1: Bell felt the need to ask one of his Vegas "experts" how much home-field advantage is worth in the World Series.

This was unnecessary to begin with, since the value of HFA in a seven game series easy to calculate. MLB home teams win about 54% of the time, so all else equal, the home team will win the series 51.25% of the time. Going from having HFA to not having it is about a 2.5% swing. This can be calculated by doing some conditional probability in Excel.

Concern #2: Supposedly, his Vegas "expert" told him HFA is the difference between being 57% to win the World Series, and being 45% (a 12% ASG swing!).

It's not entirely clear what happened here. Kornegay is on the record as not being math's biggest fan, but this is such an insane answer that that doesn't full explain it. Maybe he said 47% and Bell misheard him? Maybe he just pulled a number out of his ass because he doesn't care? The world may never know. Regardless, after this tweet it's hard to put a whole lot of stock in anything Bell reports as Kornegay's "expert" opinion.

Concern #3: Bell excitedly published this.

It seems like this couldn't have been a typo, since he confirmed the enormous difference with the instant classic, "12% All-Star game swing!" Anybody with any clue would've looked at this and wondered if it could possibly be correct. Bell looked at it, saw the eye-popping 12% number, and got excited about how many retweets it might get.

Concern #4: This all happened over nine months ago. Bell was called out repeatedly for how wrong it is, but he doesn't seem to care.

Accepting that he posted it in the first place, I'm not sure exactly when Bell should have fixed or deleted his insane tweet. Maybe after this polite correction:

This reply would have worried me, personally:

He probably should have figured it out after seeing this back-of-the-envelope math:

He was really given an opportunity here:

And again a few weeks ago:
Still nothing.

The combination of incompetence and complete disregard for the truth is the most frustrating part of all this. It's absurd enough to tweet out something so phenomenally wrong in the first place, but to leave it up with no correction really takes it to another level.

Although I guess I should be careful what I wish for, given the alternate option:

Pretty worthless tweet to begin with, before you get to the part where it's completely wrong:

This time, Bell eventually realized his mistake. Didn't correct his error, but did get 75 more RTs on game day:

Bell also thought that going from 64 to 68 teams increased the number of NCAA tourney bracket possibilities by a factor of four, which is not how math works. He acknowledged his error in that thread, but not in the actual article (see bobderwin's comment), and not before multiple other outlets picked it up.

Despite what he'll tell you, Bell is not knowledgeable when it comes to sports betting markets. He is not good at math. He has no interest in taking accountability for his mistakes unless he's left with no alternate option. And on top of all that, his incentives are all screwed up, because at the end of the day he just wants to get you to his tout site and sell you some shitty picks.

Given all that, why would you ever want to rely on him as a source?

Previous Long Con post: World Openers
Next Long Con post: #TERNDZ


  1. Surprised you didn't talk about his inability to calculate vig free odds...

  2. "I love the theory talk as much as the next guy, but you need to communicate in broad strokes with the general media. In the USC game, -210/+180 was widely available, the juice free line = 33.8% . . . but the media doesn't like "decimal point thinking" . . . I've seen guys calculate from the fav lay, which would be 32.2% (not an approach I'm a fan of, btw) . . . the key is to convey in an easy to understand fashion for the general reader, and that's what I do with wide releases." via

  3. Đồ nội thất hòa phát là sản phẩm phổ thông và được sử dụng nhiều nhất trong tất cả các văn phòng, trường học, gia đình,...Mỗi sản phẩm của hòa phát đều được gắn thương hiệu trên từng sản phẩm. Hiện nay trên thị trường có rất nhiều cửa hàng treo biển bán sản phẩm nội thất hòa phát với giá rất rẻ. Vì vậy khách hàng thường nhầm tưởng giữa sản phẩm nội thất hòa phát chính hãng và sản phẩm nội thất trôi nổi. Cùng chúng tôi đến công ty mua bàn làm việc nhân viên chính hãng có đảm bảo chất lượng và bảo hành sản phẩm cho khách hàng. ban van phong hoa phat có rất nhiều kiểu dáng khác nhau, cùng phối hợp các màu sáng đẹp, tạo nên một sản phẩm bàn làm việc hiện đại. Bàn làm việc văn phòng gỗ được sử dụng nhiều nhất là mau ban lam viec van phong có hộc CPU rộng và ngăn bàn để bàn phím máy tính, ngăn bàn nhỏ đựng tài liệu và 1 ngăn kéo ra kéo vào đựng tài liệu quan trọng, hay các loại bàn làm việc chân sắt với thiết kế đơn giản, nhanh gọn nhẹ vói chân sắt sơn tĩnh điện mặt bàn phẳng rộng thích hợp cho mọi văn phòng làm việc.

  4. Deetzmon here, strangely his odds had the right numbers but in the wrong order 31200 vs 23100