What is ‘The Edge’?
Last updated: Aug 22, 2019, 3:15AM | Published: Nov 22, 2018, 1:57AM
You may have seen ‘The Edge’ on the TAB betting website, app, and TV ads. Something you may have missed, though, is that ‘The Edge’ is powered by Stats Insider.
In fact, ‘The Edge’ IS Stats Insider.
So, what exactly is ‘The Edge’?
Let’s start at the beginning.
HIGHLIGHTS:
- Stats Insider identifies value in the betting market using advanced modeling techniques, as opposed to being your standard ‘tipster’;
- A smiley icon is not a confidence rating, nor a ‘tip’;
- Probabilities have the ability to vary over time, based on changing information and betting trends;
- The model is trained on - and aims for - long term success against the betting market. Short term volatility is inevitable.
Stats Insider predicts the probability – the mathematical percentage likelihood - of a sporting result or outcome. We have built our own machine-learning models powered by thousands of data points, to help punters get an edge over the bookmakers.
How does Stats Insider work?
After all the data has been sourced, it goes into a custom-built database. Our machine-learning algorithms vary by sport, but they essentially “simulate” the match, race, or even full season or tournament. We simulate the result of each event 10,000 times, in order to calculate the probability of a certain result or outcome occurring.
From our 10,000 simulations of an event, if a result occurs 5,800 times, we show that result as a 58% probability. It also means that in our simulations, this outcome does not occur 4,200 times, or 42%.
So, what is ‘The Edge’?
An edge is how we display the difference between our probability of a result, and the bookmakers implied probability of that same result, ie; the odds.
For example: a Line/Spread market is – at least initially - a 50/50 market, in that the odds will be equal on both sides of the spread, usually around the $1.91 mark each, so the implied probabilities for both sides would be 50%.
DID YOU KNOW?
Our algorithm is predictive, not historic.
You might hear or see tipsters using ATS/BTS records in their tips.
While we use a lot of historical data in player and team performances/results to train our models, the fact one team has won eight of the last ten versus a particular opponent, or that a particular player has been Man of the Match three games in a row, has very little to no impact on our predicted outcome of the next match.
On Monday 12th November, the NFL’s Atlanta Falcons played the Browns in Cleveland. The Falcons – Superbowl contenders just two years earlier - with a high powered offense and MVP calibre Quarterback Matt Ryan, were favourites in this game, with the market spread being five and half points, Atlanta’s way.
This table (our match page ‘Value Summary’) shows the difference between bookmaker probabilities (using Average Market Price – odds - displayed as a percentage) and the probabilities generated by the Stats Insider model across popular betting markets, after running 10,000 simulations of the match.
So how does it work and what do I do with it?
The Stats Insider model has Cleveland to cover the +5.5 point spread in 61% of it’s 10,000 simulations, compared to the bookmaker’s 50% market. Based on this data, our edge percentage has identified that the odds for the Browns to cover the +5.5 line should more likely be around the $1.64 mark, so if you can get the odds at $1.91, that’s a clear betting advantage. This is displayed by the "Green Smiley" icon which indicates that betting these plays offer a long-term, profitable betting strategy more often than not.
That 11% differential? That, ladies and gentlemen, is ‘The Edge.’
As you can see in the Value Summary above, one market (Line) has a green smiley face icon next to it, one market (H2H) has an orange smiley face, and one market (Total) has the grey smiley.
This is how we display whether the model has identified betting value in a game or not. A green smiley indicates that based on the model’s training for long-term success, this looks like a high value play to make. An orange smiley indicates this type of play as essentially breakeven or slightly profitable long-term, while the grey icon indicates a “no value bet” for this particular market.
You’ll notice the orange smiley on the H2H market . Across bookmaker markets, the Browns were, on average, around $3.33 underdogs in this game, shown by the 30% probability of a head-to-head victory in the Average Market Probability column. The Stats Insider model identified a 6% edge for a straight-up win for the Browns at 36%, where the implied odds would be about $2.77. While this represents value compared to the bookies price, our long-term profitability analysis suggests that this bet is likely to be only slightly profitable in the long-term, compared to the Green Smiley 11% edge play on the Line market.
What is important to remember is that the smiley faces are not an indication of betting confidence. What they reflect is the long-term success rate of similar projections, and we believe that over time the projection will show a profit against the stated betting market.
Of course, we now know that, not only did the Browns cover the +5.5 point line, they also won the game, putting Stats Insider's followers at a clear advantage over other punters, having identified value on the underdog in this game.
Why do the predictions change over time?
Our predictions are dynamic and completely automated.
This means the projections can change as new information is gathered since the last update. The models are also trained on movement in the betting markets, so probabilities can change the longer the market is open, or following a big swing in the market which may represent information not available to the model.
Should you always just do what 'The Edge' says?
The model has no human emotion. As punters we’re all used to looking for that great piece of inside information that gives us a betting advantage. So, while the model is more likely to find value against the market in general, it won’t know if a stadium’s field is more waterlogged than usual, or torn up due to a concert, whether the starting point guard is going through a particularly harrowing personal situation, or something else that has the possibility to affect performance outside of the standard variables. This human knowledge still has a place, so use your own knowledge in combination with our data analysis for a solid advantage.
You might also notice that in one particular sport, a market with a 6% edge may display a green smiley, whilst for another, an 8% edge might not show a play at all. The smiley thresholds are based on long-term analysis of value propositions in each sport, and the threshold for recommended plays differ from sport to sport; based on projected and historical results, and the size of the market (e.g. EPL and NFL are more efficient markets than, say the NRL or AFLW).
The reality is - an edge, is an edge, is an edge – whether a smiley face is displayed or not, it is completely up to you whether to take the value in the model’s projection. The smiley faces are just our suggestions based on our long-term analysis and results.
We do preach caution with the odd market that may show an extraordinarily large edge. If a projection seems exceptionally different from the market compared to other edges found in that sport, it may mean a crucial piece of information is missing – for example an injury rumoured on Twitter that is not officially recognized in our data feeds yet.
The bottom line here is that some sports are more profitable than others, due to the efficiency of the betting markets available, and the sharpness of bookmakers, let alone the smart punters. But, that’s another story for another day – stay tuned to Stats Insider for more valuable play-making content.
Did you enjoy this article? Leave a comment below, or join the conversation on the Stats Insider Twitter or Facebook page.