Stats Insider is empowering Australian sport. Our vision is to evolve the sports fan experience by making data-driven content widely available to sports nuts like us. Launching in 2018 with AFLW, and since publishing predictions across over 15 sports codes, we are dedicated to leading the way on advanced analytics content and discussion in Australian sport.
Built in-house at our HQ in Collingwood, Stats Insider’s machine learning models take a number of different approaches to predicting the outcome of sporting events. In general, our models are based around using a Monte Carlo approach to simulate each match 10,000 times to understand the range of outcomes, and how likely each of them is.
In most sports, such as NRL and AFL, we first predict the performance of each individual player, based on hundreds of algorithmic factors for each sport, and then use this range of player performances to run our match simulations. If a team wins 4,000 of our simulations, we project them to be a 40% chance to win the upcoming match.
Our models update throughout the week, and take advantage of all the information available at that time: team lists when they’re announced, weather conditions when they’re known, moves in the betting market as they happen. When all is said and done, you’ll use your own knowledge and the information we give you to engage with the sport - whether that’s as a viewer, as a punter, or in your office tipping competition.
Our models also predict the outcome of major thoroughbred races, and we offer an easy-to-use form guide for your days at the races, at the pub, or on your sofa. That form guide includes our value assessment, predicted win/place % and deep-dive analysis for each horse, all of which utilises our machine learning models, with the win/place %’s being generated through a Monte Carlo approach as described above.
We are driven every day by the need for intelligent sports coverage and bloody good data across the codes we love in this country and hope that your experience with Stats Insider exceeds expectations. If there is something we could be doing to improve your experience, or if you see an alignment between what we do and an idea you have, please make sure you reach out to us so we can get talking.
While everything on the site is free to view, we encourage you to register a free account in order to participate in active conversation across articles and match pages.
The Stats Insider punting information displayed on selected pages across the site is designed to give our users an edge against the bookmakers. This information should only be used as a guide to navigate the best value plays on upcoming matches. Here are some important things to know if you plan on using this information:
We’re not a tipping service.
Stats Insider is closer to a form guide than a tipping service because our predictions move with the market and with the information known, so the investment suggestions change based on the best value currently available. We provide the information 24/7, you decide what to do with it.
This is what we do.
We’ve been building, developing and testing our models for 15 years, and continue to add to and improve them daily to bring you the best possible predictions every time. It’s our sole focus and what we dedicate all of our time to. There is a lot of information out there, and you can build your own models to take over the world, but chances are you don’t have the time to do that for every game in every sport. That’s why we’re here.
Predictive, not historic.
The fact that Geelong has won their last 4 out of 5 games against Melbourne does not necessarily mean they are going to win the next one. What if 3 of their key forwards are out next week, what if they’re playing away and in the rain? Historical stats are not a predictor of future performance. Predictive analytics take these things into account, and much more, to give you a true prediction of how a game is most likely going to play out, based on 10,000 simulations of the upcoming match-up. Read more about how we build these predictions here.
It's easy to find & free.
The Model Value Summary page is the fastest way to see all of our value plays for upcoming matches. You can also find detailed information on each value play on the match page.
We believe data-driven wagering intelligence is paramount to decision making in betting, rather than relying on emotion or gut feel. By making this information completely free to access, our goal is to help punters make more informed investments, and ultimately get an Edge on others in the market, wherever you bet.
Beat the bookies while they’re behind the ball.
It’s going to take the bookies a while to catch up with the times, and by using Stats Insider, you have the ultimate head start to catch them out. The trick is timing – use Stats Insider predictions early in the week, or when predictions go up, to get ahead of the market.
We bet on our own predictions because they win, but we’re not idiots. If it’s not a good bet, or if the risk is too high, we give it a miss and wait for the next bet to pop up. Betting smart means higher profit margins and more money in the bank for the next punt.
Ask and you shall receive. Below is a constantly updated list of questions you have asked and we have answered. If we haven’t covered what you’re looking for below, submit a new question here.
What is your ROI?
Because our projections move with the markets, it’s impossible to provide a fixed ROI. If you’re betting on a Tuesday night after NRL teams are released, you’ll get very different suggestions to a Sunday afternoon just before the game, but that doesn’t necessarily mean one is better than the other. We do regularly publish content showing results if you bet at different times, but these are simply guides against a single bookmaker — you should always shop for the best price.
How can you pick the underdog to win but the favourite to cover? Shouldn't they match up?
We use match-specific push percentages to convert between a H2H and Line pick. Occasionally these will be out of line with the bookmakers method of converting between them, which is why you may see different selections for H2H and Line bets. In particular, this can occur when we’re expecting a significantly lower or higher scoring game than the bookmakers, or when our simulations suggest a team is highly volatile.
What time are your picks produced? Where do you get your lines from?
Initial projections for a week are released within 12 hours of lines becoming available, or 24 hours after the final game of the previous week. They then update at regular intervals throughout the week, taking into consideration updated information (team lineups, weather etc) and market movements as they become available. We source odds from multiple leading Australian bookmakers, to cover the wides array of markets possible.
How should I use these picks?
That’s up to you. We bet selectively, and roughly quarter-kelly according to the edge of each bet (as described by the smileys). There will be many games (most, even) we don’t bet on. But, you do you. If you want a $10 bet on the game because it’s on TV, go right ahead.
How do you deal with lineups/injuries/benchings etc?
Our models are all player-based, and take account of the current projected lineup for the game. You may therefore see significant updates when a team is announced (e.g. Tuesday for NRL, Thursday for AFL), and again once final teams are known in the hour before the game.
Stats Insider had a Eels +7.5 bet on the early lines, why has it disappeared now?
Our models run every 3 hours; and take in all the information available at that stage (e.g. weather changes, updated teams, line moves etc). We differentiate from a tipping service in that way; the goal is to have the most recent/updated information reflected in our predictions for the match; and the smiley faces are a visual representation of the value at a given point in time.
For example, we might publish an NRL early lines article on a Broncos/Eels game early in the week, but the match page will shows a different prediction to the early lines article as early as 3 hours after. Let’s say the Eels line moved 7.5 to 6.5, and the total also moved, the combined effect of which took it below the threshold for a 1-star bet (value bet). Eels +6.5 was rated a 55% chance at kick-off, and the threshold for 1-star is around 58-59%.
Why does Stats Insider only have AFL value bet?
This is the only bet with an edge at this point in time, based on our value bet star rating. This might change as the lines/totals move and further team news comes through over the weekend leading uptown games, however. We only show a value bet if there is a 58-59% chance of it happening, considering all of the information that is known at that time. For example, the Western Bulldogs +18.5 might be at a 56% chance before teams are known, but once the model updates with team news an hour before the game, that bet might go up to 60% and become a value bet against our star rating.
How do our predictive models work?
Each of the Stats Insider models uses a fairly similar approach to simulating a sporting event. They vary in complexity and approach: our NRL & AFL models are more sophisticated than our AFLW ones, because there is more information available. Our Cricket model differs from our ‘ball-sport’ ones, because Cricket is effectively a series of individual matchups between a bowler and a batsman (with some fielding quality impact) where as ball sports are impacted by the ability and positioning of multiple players at a time.
In essence though, we seek to predict the distribution of a players performance (e.g. how often a player will score 0, 1 or 2 tries, how often they’ll have 10, 20 or 30 disposals). Once we have those distributions, we use a Monte Carlo approach to predicting the outcome of the match; that is we take those player distributions and simulate the game 10,000 times, so as to understand how likely a team is to win, how likely there will be 40 points scored, or how likely a player is to score the first goal.
Once we’ve run those simulations, then for the punters and tippers amongst you we can compare the results to common betting markets, and the odds on offer. If a team wins 6,520 of our 10,000 simulations, then we predict them to be a 65.2% chance of winning the match. If you’re able to bet them at $2 at your local TAB or friendly online bookmaker, we might highlight them as an investment worth considering further.
How accurate are they?
If we’re predicting a team has a 55% chance of winning, or even an 80% chance of winning, that’s not saying they are going to win, or even we think they’re going to win. We’re saying we think if they play this match 100 times, they will win 55/80 of them. That’s a crucial difference, and underlies everything we publish on Stats Insider: we provide data, not tips.
We won’t publish anything on this site that we do not believe to be profitable against available odds, and similarly everything we publish we expect to be within a reasonable margin of error in predicting match or season outcomes. That said, the profitability of the models will vary from sport to sport and market to market, and we would expect to lose 45-46% of our line or total bets, and to have losing periods – and even seasons – across all sports.
Why do they change?
If you came to the site on Monday and we highlighted the Broncos as an investment worth considering, but by Saturday we’re suggesting you consider opposing them, don’t be surprised. Our models gain information all the time: on Tuesday when (NRL) team lists are announced, on Thursday when it’s announced the hooker has picked up an injury, on Friday when one of the big betting Syndicates hammers the price, and on Saturday morning when the BOM issue a weather alert for a severe thunderstorm. All of those things impact the possible outcomes of the match, and so all of them impact our models. Our suggested investments may change frequently right up until game time.
How do your in-game simulations work?
They’re very similar. 10,000 simulations take a fair bit of time – and computing power – to run, and we can’t do that every minute during the game. Depending on the sport, we may run 500 or 1000 simulations of the rest of the game. But the approach is the same: if we played the rest of this game numerous times, how many times does each team win?
How about the projected ladder?
Our seasonal projections take two primary inputs: the individual match simulations (that is, we simulate each match remaining in the season 10,000 times), and the likelihood that a team will be impacted by injury to one or more of their key players. If a player has historically missed 25% of the season through injury, or if they’re in doubt for the coming week, our season simulations will capture this and force the team to ‘play’ some of those games without that player. Our seasonal simulations are more complex, and take longer to update, so you may notice these lag a day or a few days behind the individual match projections.
What should I do with the information?
Well, that’s up to you. We bet many of our suggested investments each week. We enter them in tipping contests. We use the information to inform our fantasy decisions and pick our daily fantasy teams.
But, most of the time, we just use them to watch the games from a different angle. At kickoff, the Tigers don’t have a 50% chance of beating the Lions, it might be closer to 70%. When a referee makes a controversial decision and disallows the try, we can see that 15% increase in win probability come right off the board. It makes you appreciate every play that happens. It makes sport more fun.