Tom Jessiman, the founder of PicksPal, stumbled onto something that could change the way we predict sporting event outcomes. And it could therefore have a big impact on sports betting markets.
PicksPal is a free sports site where people “bet” on upcoming games. No money is involved. If they win, their point total goes up and they have bragging rights around the office. Since launching about a year ago over 100,000 people have joined the site, making daily picks on just about every kind of sporting event in the U.S. – boxing, NFL football, pro football, bass fishing, ultimate fighting, basketball, baseball, etc. The site makes money from advertising.
Recently, however, the PicksPal team noticed that a very small percentage of users tend to be correct in their picks significantly more often that they should be statistically. When they grouped these special users they found them to be a powerful predictive force.
Sports betting (both legal and illegal) is a massive worldwide business. Participants are always looking for an edge and are willing to pay for picks by “experts” (I know this from my extensive research watching Two for the Money).
Tom and his team figured out pretty quickly that there was a potentially massive business here. And next Tuesday they are going to start tapping into that business when they launch what they call “Genius Picks”. For $10, users can get access to the collective wisdom of the 30 best PicksPal players over the previous five weeks in a given sport, and get five predictions on upcoming games.
They’ve been running this internally for a while and the results are impressive. This last weekend, for example, they went 5-0 against the spread, picking Toledo -3 in college football, and Indianapolis – 13, Cincinnati – 10.5, Baltimore -9.5 and San Diego -11 in the NFL. Tom says not to expect Genius Picks to maintain that kind of record over the long run, but the results speak for themselves.
PicksPal will be leveraging their technology to participate in completely new markets later this year. They say a “Hollywood” version could be next, to allow people to bet on who’ll be on certain magazine covers, reality show plots, etc.
Look for mainstream media to start covering this company in the very near future.
PicksPal is based in Silicon Valley and has raised $6 million in venture capital over two rounds of financing. They are backed by Canaan Partners and Bay Partners.
Wasn’t the CEO of an online betting company arrested recently in US?
no idea. Tom, the CEO mentioned in the post, was previously the COO of CBS Sportsline.
Top executives of both BetOnSports and SportingBet have been arrested in the US in the last few weeks/months. Online sports betting and the facilitation thereof is illegal in the US based on the highly outdated and over-protective Wire Act.
Think these boys should start moving their servers to the bahamas.
unbelievable. it seems very, very cool, but it also seems like something’s missing. is it possible that nobody in the gambling business thought of this before? or maybe this is just the first time it’s being done online, thus enabling a ’super batch’ of top predictors?
I never bought into the ‘wisdom of crowds’ thing, and for the same reasoning I _have_ always bought into the ‘wisdom of extremely competent crowds’ thing. It’s part of why I respect Google.
i would have to think Wall Street has been doing this for ages, but I remember seeing some tv stuff about different ‘lone superstar’ mutual fund managers, who controlled funds of ‘x’ hundred million dollars and all that. I never got the impression that these solo genius fund managers had their ‘wisdom’ checked by others to build a consensus – and too often they seemed to lose money or not even match the DOW in growth, etc. (even with their power to influence the market, insider knowledge, etc.) Forget Hollywood and sports betting – someone on Wall Street is gonna want this built out. I know there are all those ‘wisdom of crowds’ sites in the financial arena, but I guess they just got beat to the punch on this one. Should be interesting to see what happens.
wow. if these folks could be right, say, 60% of the time…
and there’s always the possibility that they could be outrageously consistently correct (70%+) in predicting some very specific types of opportunities. Like, how are these folks predicting Steelers, regular season, home, against non-divisional opponents, in October?
Wisdom of crowd ideas have been used for a while in totalizer based horse racing events. William Ziemba published a number of papers showing how the win market for a race was a good predictor of the actual chances of a horse winning. Those runners with 30% of the money bet on them in the win pool tended to win 30% of the time. So the wisdom of the public as generally good at rating there chances of winning.
This is no good for actually making money on the win pool though since it’s a tote – it’ll never pay more than the percentage, combined with the totes rake you lose. But there are other pools on the same race. Show and Place pools where you bet that the runner comes 1st, 2nd or third. Or trifecta and quinella pools.
It seems the public is good at computing the winning chances of runners but lousy at picking events like place, trifecta, etc. But given the win % of the runner it’s easy to compute the chance of runners placing, getting the trifecta, etc. So given the knowledge of the win % predicted by the public as being accurate you can look for ‘good bets’ that statistically will turn a profit over time in the other pools.
That’s the theory anyway – there are difficulties in putting it into practice but it does demonstrate a use of the wisdom of the crowds in gambling in the past. Google for ‘Dr Z.’ or ‘William Ziemba’.
Chris.
Brilliant, with a capital B.
This looks really cool. As far as the legal issues are concerned, the founders must have looked into it for sure. But the market they have identified is brilliant. Lets see how does it do in the long run.
This is food for statisticians and economists. It extends far beyond the limits of betting as it may apply to any situation where a future outcome has to be predicted.
Just wondering how this may work in the financial arena. Evidence repeatedly demonstrated that there is no way to consistently beat the market on the long run.
Or may be it’s possible if the edge in knowledge/capacity is concealed from the market.
It is perfectly legal to sell picks as Wayne Alen Root (google him) and others have been doing for years. Selling info has never been illegal as far as I know. Gambling is however unless you’re in Vegas or on an Indian Reservation or off shore. Using the internet to place wagers on off shore servers is what the govt. is trying to stop using a 1960’s act intended from limiting mob bookies. Those arrested recently were running sloppy operations and actually took wagers over the phone from US locations.
Sportsbooks make money by adding a vig (vigorish) around 4% to the wager. So, to keep people betting on both sides of the game they give the underdog points. They have got to offset the bets. As a bettor, you are not betting against the house, you are betting against the wisdom of the crowd. Which, is how the system has always worked. Now, it will be wisdom of the crowd vs. wisdom of the crowd.
Vegas (sportsbooks) has known about the wisdom of the crowd forever. Handicappers (those who sell picks) have always looked at where the crowd has misjudged. There will always be people who wager with their heart and there will be those who use statistics and try to pick with their heads.
Handicappers, who sell season picks (educated guesses) for upwards of $700, will be the only ones disrupted by this. It’s a brilliant idea.
Bottom line – The human element is why there is never a “lock”. Just look at the Eagles last week!
My guess is that the top pickers will start to exit if they don’t get compensated from the revenue – do they plan to pay these people? If I’m 70% right – why would I let these guys profit from it?
Incidentally, there have been several other plays like this in the past – HSX.com comes to mind – and there was a stock-picking service that I can’t remember the name of that employed the same methodology. Meaning, people built virtual portfolios and then they built a total portfolio based on the top pickers – if I remember correctly they did really well for a while and then it blew up.
I smell problems.
1) “For $10, users can get access to the collective wisdom of the 30 best PicksPal players over the previous five weeks in a given sport, and get five predictions on upcoming games.”
Statistically speaking, approximately 3,125 of the 100,000 users would have got the past 5 weeks of predictions correct through pure chance (100,000 x 0.5^5, ignoring draws admittedly).
If it’s NOT the same 30 PicksPal players each week giving recommendations, then a lot of them have got there by pure chance.
2) If these guys do have special powers, then by selling this information to the market, they will eventually kill any potential returns, as the odds change to reflect demand/supply.
3) “Recently, however, the PicksPal team noticed that a very small percentage of users tend to be correct in their picks significantly more often that they should be statistically. When they grouped these special users they found them to be a powerful predictive force.”
It would be weird if a very small percentage of users were NOT correct in their picks significantly more often than they should be statistically.
Either the premise of the above is that these users have some special predictive powers, in which case, Pickspal are very generous to sell this at $10 a pop(!) as they could make much more amongst themselves.
Also, if these group of users aren’t fixed permanently, then there is nothing special going on.
Read Fooled by Randomness by Taleb for further clarification.
There are sites like this in the UK that have been operating for years. Nothing new here.
#10 — I agree with damian.
I think they should fit the business model to fit the reason this second idea even exists in the first place.
Did anyone leave digg for netscape’s proposition?
Why somebody would pay 10$ for collective wisdom, if there is no money involved in his bet? Just for fun to win? Or this site information will be used in real money bets ?
The reasons for those arrests were for more then just running a gambling site (which I think was already over seas). They were charged on racketeering and evading taxes. It is not illegal to sell sports tips for a flat fee.
This has been done on the market side of things at http://www.marketocracy.com
They perform “okay.”
MAxim, the picks will be used with a bookie for making real money picks afterwards which is illegal, but either way it goes on anyways. There are magazines out that pretty much do the same thing as this website, showing spreads and possible bet schemes. Seems like nothing new to me, except they also offer the free site to bet points instead of money.
I like the idea.
Can MoveOn (#11) share which sites in the UK offer this already?
I really like this idea, I have a great interst and passion for FA Premier League football but I am biased towords Liverpool FC and its great to get some info from other supporters and gamblers that know their teams best! Win- Win situation!
TradeSports.com has been in the prediction market (for money) for years. I’m certain they have a great deal more data than PicksPal.
This seems like one of the more interesting companies featured on TechCrunch as of late. Even if this isnt the first of its kind, Innovative ideas like this are great.
I dont know about the rest of you, but I am starting to get tired of reading about a new photo sharing company or a new online storage company popping up ever 23.5 seconds!
“They’ve been running this internally for a while and the results are impressive.”
The site continues to sink lower. Now you are supporting a site that claimsto give sports bet winners. … Holy internet 1997.
The title to this could not be more overstated.
sports gambling is a self correcting market. if too many people bet one way on a game, the point spread starts to shift. bookies want people to bet evenly on every game and just make their 5% or so margin that they charge. How much people win week after week is pretty irrelevant as long as someone else is losing just as much.
and if the picks really were that good they wouldn’t be sharing them with the world. they would be placing bets themselves.
Maxim are you joking? Are water bongs for tobacco use only as well?
So what happens if lots of people start using this? Sportsbooks operate like stock markets. The odds at sportsbooks are changed to reflect incoming money, so if lots of people use this data it will be self-defeating. Long-term this may provide decent returns, but every other system like this (like Marketocracy) never has unheard of success over the long-haul. The smart money just doesn’t share this info either.
(1) You can be correct in your picks significantly more often than the average – always bet on the favorite. Keep your $10.
(2) The people in that best 30 will be those who got lucky in the last five weeks, not the best, because although there are real experts with an “edge” it’s going to be a few percent and not enough to outweigh random dumb luck in the short term.
(3) Why aren’t the founders of PickPal just keeping this advice secret and betting their own money? Why do you think?
let’s see what they do at the end of the season. look at their 5 plays this week. 4 big favorites and one small favorite that actually covered in OT (lucky play).
If the top selections are chalk every week, then at the end of the year, these people will be losers. I think only 3 NFL dogs covered this week, so Vegas got hammered.
Call me skeptical but it is easy to retroactively look at the hundreds of predictions made and then choose the small subset that proves your theory — then say that it has “predictive” power.
I’d like to see what the genius picks are for this weekend published in advance here. Let’s then check back and see how they do…
“This last weekend, for example, they went 5-0 against the spread, …Tom says not to expect Genius Picks to maintain that kind of record over the long run, but the results speak for themselves.”
What exactly does a 5-0 record in a single weekend say for itself? Wagerline (http://www.wage....aspx?sportid=1) has been giving out their top performers picks for years and it hasn’t disrupted anything.
If 10,000 people all bet on heads/tails of 10 coin flips, a decent number will get all 10 right. Are they smarter than everyone else? Can I start a site that gives you access to their picks for the next 10 coin flips?
http://en.wikip...rvivorship_bias
old idea
very very old ones
http://www.asianbookie.com is still the top
Very cool idea! These guys are selling the wisdom of the smart predictors. Listen to ESPN on Sundays and Chris Berman makes picks, he’s just not very good. Nothing illegal with that and no betting associations. I think the idea of moving this to other markets like W Street and Hwood is pretty neat. Although with Hwood the thing would be to watch for insiders. Agree that this is one of the cooler ideas on TC lately. Keep it up!
This idea is interesting, but it could be refined a bit. 5 weeks in a given sport is hardly enough to validate someone as an “expert”. You need a larger sample size to get a better gauge of their abilities (assuming that there is some skill to this, and it’s not just all luck). Beyond that, they should refine the expert picks further. For example, they could follow someone’s track record with a certain team, within a certain conference, during a certain part (beginning, middle, end) of a season. Another good thing to have would be a pick confidence heuristic… you could show how confident the “experts” are with a pick by weighing the total picks intelligently.
Sorry, but nothing will disrupt sports betting markets.
This is less usefull then following the “Smart” money.
re: previous comment from Paul about Marketocracy’s “okay” performance, judge for yourself witih real data:
Here are the top 10 fund managers in the community with the best 5-year track records:
http://snipurl....ketocracy_top10
and performance of the actual fund versus the S&P500:
http://snipurl....tocracy_v_SP500
Predictit.com had this exact service in the late 90s / early 2000s. It’s hard to find information about them because rather than disrupting the sports betting markets, they simply folded.
this website is not useful, and people do not pay for picks unless:
someone has inside information
someone handicapping games has had years of success (every year) doing so
there is hard data to back up the pick
-just throwing a bunch of picks out there and expecting people to pay money for them is never going to happen. yes it’s $10 but even a $500 bettor won’t waste time with this. Obviously these VC guys have no gambling experience.
Predictit was a pig. This site is a lot more interesting. I can see lots of people using this just as a kind of office pool. Think about March Madness and there’s something like 100 million people in the country who play that. If these guys could do something with that that would be very cool.
this site is great! I am addicted already. you can even do parlays. sweet.
I think it’s just ramdomness. In a group of 100 people, 50 will choose an A/B choice correctly at random. The next week of those 50 that chose correctly, 25 will choose correctly. The next week, of those 25, 12 will choose correctly. By the fourth week, we are down to 6 peope who have chosen correctly all four weeks, but it was all just random. If I sell the choices of those 6 people for money to people based on their idea that they have a greater chance than random of choosing correctly in the fifth week, I am misrepresenting their abilities. It’s still just random, and 3 of those 6 will chose the winner and 3 will chose the loser. But then someone will claim, “But those three that are left must be really, really smart — they got it right all five weeks!”
I’ve been a member of PicksPal.com since March madness and I’ve loved it. The odds are legit lines, they cover all sports, and the props make it fun. A true test of your sports knowledge without breaking the bank, and you win prizes. Unbelievable users on the site, who are great to challenge in leagues and one on challenges.
This Pickspal site is legit. I joined about 5 months ago.
I’ve seen them on fox, on the radio, and read some positive articles in the newspaper about their past success.
Its a fantasy game where you compete to win prizes. espn offers similiar contests and get great info.
All these guys on tv think they are good and try to make predictions, but are wrong 60% of the time. Lee Corso was 4-20 at one point last season.
This site lets your picks do the talking and best of all its FREE to play. We have a league here at work and everything is updated live and real time. WHere else can you pick horses, NFL and college football all day on props and games without any money or monthly fees.
@Gaius
Funnily enough, what you just outlined is how most of the major professional picking services work….
They just garuntee wins every week, and cold call like crazy to get people to get their “free” pick. Odds say that the company will get all the picks right every 4 or 5 weeks. So on the weeks they are wrong, they lose all the customers, on the weeks they are right they are able to retain a bunch of people who go “oh wow! They just went 5 for 5!!!!” and then pay the company a few hundred bucks for their services for the rest of the season. In the end they generally end up being no better than just picking yourself.
The hook of course is just to nab enough people every 5 weeks with your “perfect record” that you make money. Everyone else from all the weeks you just let walk away, save their phone numbers, and try them again in a few years after they’ve hopefully forgotten that you actually suck.
Surely these chaps, knowing they are in the top 30, should quit playing for free and go down to their local bookmaker and make a fortune?
It’s been statistically disputed that experts are any better than backing the favourites, in fact, typically much worse. Follow the money and you’ll do better.
I’ve heard of Pickspal on the radio. Some x NFL Player and their chief handicapper who did real well in the Las Vegas Hilton NFL Contest last year hitting 64% throughout the entire NFL Season. They do a weekly show and daily Pod Costs breaking down the NFL and College weekend games. The site is tough to beat, because they have so many players, but their predidctions do well.
I think its a fantasy website not a sportsbook or casino.
No other site out there like it! You can play sides, totals, and props on just about any sport in the World. It’s a free fantasy style game that could win you some nice prizes.
Jeff
This is a great site for fantasy sports players i like playing the fantasy sports but this also lets you picks the games themselves and doesn’t cost anything which for college students is very good and its all done online so it tracks your status for you.
I find it depressing because I was part of a startup that did group and peer-to-peer “betting” using points and providing sponshored contests. The “no other site like it” and “brilliant” comments are nice to hear given that it wasn’t so brilliant 5 years ago when the investors all pulled out
Here’s a little sample from the wayback machine: http://web.arch...allpredict.com/
Mark
Clint: I’d love to show you a chart of my Insider Trading fund. It puts Marketocracy to shame. Anybody that thinks they’re going to get rich by spending $10 for “expert” picks is a fool. The real money is being made by people with insider knowledge and they’re not sharing it with the crowd.
By the way, I keep getting spam emails from PicksPal. Just received one today from some company antmorn.com. This place claims to be CANSPAM compliant but I can tell you that I never signed up for this crap. It looks like they have an interesting service but as a principle I never sign up for sites that use spam as a marketing technique. Anybody else getting these?
“# TD
September 19th, 2006 at 12:55 pm
this site is great! I am addicted already. you can even do parlays. sweet.”
looks like Pickspal trolls as well as they handicap. my prediction is they fold within the year and i’ll bet i can find 4 other experts to back that pick up.