Mail
PrintBefore the Summer Olympics began, we wrote a post asking if a Colorado College economics professor’s model that used only non-athletic data could help with betting gold medal over/unders.
The answer: No.
Hell no.
The professor, Daniel K.N. Johnson, claimed to have a 93-percent accuracy rate over the past six Olympic Games. It sure sounded great, but we were admittedly skeptical, noting, “It’s still unclear how Johnson is judging his accuracy. A 93 percent accuracy rate sounds amazing, but if he’s counting the dozens of countries where zero medals are all but assured, that’s just padding the stats.”
We don’t care to go through all the countries for which Johnson predicted zero gold medals. He probably got close to 100 percent of them correct. But, of the 23 countries that had gold medal over/unders at 5Dimes, Johnson correctly predicted three of them (13 percent).
Of the five most “marquee” countries (USA, China, Russia, Great Britain and Germany), he was off by 12, 5, 1, 9 and 8 medals, respectively, or an average of five gold medals per country.
Johnson predicted the USA to get 34 golds. They got 46. He predicted host Great Britain to get 20 golds. They got 29. He had Germany pegged for 19 golds. They only got 11.
And on and on.
To be fair, trying to predict exact gold medal counts sounds all but impossible, so we can’t really blame Johnson. We tried to narrow it down by identifying good “over” or “under” bets to see how those would fare with Johnson’s model.
OVERS
Germany (loss)
Italy (loss)
Hungary (win)
Poland (loss)
Norway (win)
Greece (loss)
UNDERS
USA (loss)
China (loss)
Great Britain (loss)
Australia (win)
Netherlands (win)
Spain (win)
Total Record: 5-7, -2.65 units
Obviously, Johnson is just having fun with a harmless model that he’s trying to perfect. Who knows, the model might have actually performed much better in the past and this year was just an outlier.
It doesn’t really matter, though, because it’s not like he was out there selling his picks on these gold medal over/unders.
But this serves as yet another example of how packaging can be used to make results sound better than they really are. It’s what pick sellers and touts do all the time. They just happen to take the next step and attempt to turn a profit off people’s ignorance.
“How can I make my results, which really aren’t that great,” they ask themselves, “sound awesome to the public and get them to buy my picks?”
* * *
USA
Over 38.5 (+115)
Under 38.5 (-135)
Prediction: 34
Result: 46
CHINA
Over 36.5 (-145)
Under 36.5 (+125)
Prediction: 33
Result: 38
RUSSIA
Over 25.5 (+155)
Under 25.5 (-175)
Prediction: 25
Result: 24
GREAT BRITAIN
Over 21.5 (-150)
Under 21.5 (+130)
Prediction: 20
Result: 29
GERMANY
Over 15.5 (-110)
Under 15.5 (-110)
Prediction: 19
Result: 11
AUSTRALIA
Over 14.5 (+245)
Under 14.5 (-290)
Prediction: 8
Result: 7
FRANCE
Over 11.5 (+100)
Under 11.5 (-120)
Prediction: 11
Result: 11
SOUTH KOREA
Over 10.5 (+115)
Under 10.5 (-135)
Prediction: 9
Result: 13
JAPAN
Over 9.5 (+105)
Under 9.5 (-125)
Prediction: 9
Result: 7
ITALY
Over 9.5 (+150)
Under 9.5 (-170)
Prediction: 10
Result: 8
NETHERLANDS
Over 8.5 (+170)
Under 8.5 (-215)
Prediction: 5
Result: 6
SPAIN
Over 4.5 (+115)
Under 4.5 (-145)
Prediction: 2
Result: 3
HUNGARY
Over 3.5 (-185)
Under 3.5 (+155)
Prediction: 7
Result: 8
CZECH REPUBLIC
Over 2.5 (-155)
Under 2.5 (+125)
Prediction: 2
Result: 2
POLAND
Over 2.5 (-140)
Under 2.5 (+110)
Prediction: 5
Result: 2
SWEDEN
Over 2.5 (+135)
Under 2.5 (-165)
Prediction: 2
Result: 1
SLOVAKIA
Over 2.5 (+155)
Under 2.5 (-185)
Prediction: Not included
Result: 0
DENMARK
Over 2.5 (+175)
Under 2.5 (-225)
Prediction: 2
Result: 2
GREECE
Over 2.5 (+220)
Under 2.5 (-290)
Prediction: 3
Result: 0
NORWAY
Over 1.5 (+120)
Under 1.5 (-150)
Prediction: 3
Result: 2
BELGIUM
Over 1.5 (+175)
Under 1.5 (-225)
Prediction: 1
Result: 0
FINLAND
Over 0.5 (+135)
Under 0.5 (-165)
Prediction: 1
Result: 0
ESTONIA
Over 0.5 (+180)
Under 0.5 (-240)
Prediction: 1
Result: 0
