About Us
Stats Insider, Australia's leading predictive analytics website, provides Australian sports fans with innovative tools and content to maximise their enjoyment of major sporting events both domestically and internationally.
Our goal is to transform the sports fan experience by providing data-driven content that is readily accessible to sports fans like us.
Since our launch in 2018, we have provided predictions for over 15 sports leagues and are dedicated to pioneering advanced analytics content and discussion in Australian sport.
Built in-house at our headquarters in Melbourne, Stats Insider's machine learning models take a variety of approaches to forecast the outcome of sporting events.
In general, our models rely on a Monte Carlo approach to simulate each event 10,000 times to determine the range of outcomes and their likelihood.
For most sports leagues, such as the NRL and AFL, we first predict the performance of each individual player, based on hundreds of algorithmic factors for each sport, and then employ this range of player performances to run our match simulations.
If a team wins 4,000 out of our 10,000 simulations, we estimate that they have a 40% chance of winning the upcoming match.
Our models constantly update until an event commences, making the most of all the available information at that time, such as team lists when they are announced, weather conditions when they are known, shifts in the betting market as they happen, and more.
In the end, you will use your own knowledge and the information we provide to engage with the sport – whether that's as a viewer, a punter, or in your office tipping competition.
Betting Information
Here at Stats Insider, you will find suggested bets and edges against available odds. This information should be used only as a guide to identify the best value plays in upcoming events. Here are some essential things to consider if you plan on using this information for your betting:
We empower users with data
Stats Insider is more like a form guide than a tipping service because our predictions move with the market and the available information. Therefore, the investment suggestions change based on the current best value. We offer information 24/7, and you decide what to do with it. This means that a team we recommended on Tuesday may not be recommended by the time they play on Saturday, although at the time, each recommendation was accurate to our model.
This is what we do
We have been constructing, developing, and testing our models for 15+ years and continue to enhance them daily to provide you with the best possible predictions every time.
Predictive, not historic
The fact that Geelong has won four of their last five games against Hawthorn does not necessarily imply they will win the next one. What if three of their key forwards are absent next week, or what if they play away and in the rain?
Historical statistics do not predict future performance. Predictive analytics consider these factors and much more to provide a true prediction of how a game is likely to unfold based on 10,000 simulations of the upcoming match-up.
It's easy to find and free
The Best Bets page is the fastest way to see all our value plays for upcoming matches. You can also find detailed information on each value play on the respective match page.
We believe that data-driven betting intelligence is vital for decision-making in betting rather than relying on emotions or intuition. By making this information entirely free to access, our goal is to assist punters in making more informed investments and, ultimately, gaining an edge over others in the market, wherever you bet.
Beat the bookies while they're behind the ball
At times, it may take the bookies a while to catch up, 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 they go up to get ahead of the market.
The Data
How do Stats Insider's predictive models work?
Each of Stats Insider's models use a similar approach when simulating a sporting event. Essentially, our data analysts aim to predict the distribution of a player's performance, such as how often a player will score zero, one, or two tries in the NRL, and how often they will have 10, 20, or 30 disposals in the AFL.
Once these distributions have been predicted, a Monte Carlo approach is used to forecast the outcome of the match. This involves simulating the game 10,000 times using the player distributions to understand the likelihood of several outcomes, including how likely a team is to win, how likely there will be 40 points scored, or how likely any given player is to score the first goal.
Once the simulations have been run, Stats Insider can compare the results to common betting markets and the odds available for punters and tippers. For example, if a team wins 6,520 out of 10,000 simulations, then Stats Insider will predict them to have a 65.2% chance of winning the match, which is fair odds of $1.53. If the team can be bet on at $2 with a bookmaker or on an exchange, then this would be listed as a suggested bet at the time.
How accurate are Stats Insider's predictions?
If Stats Insider predicts that a team has an 80% chance of winning, it does not mean that they will win or that Stats Insider thinks they will win. It means that if the team were to play this match 100 times, the given team would win 80 of them. This is a crucial difference and underlies everything that Stats Insider publishes: we provide data and insights.
Why do Stats Insider's predictions change?
If you visit the site on Monday and it highlights the Broncos as an investment worth considering, but by Saturday it is suggesting opposing them, do not be surprised. The models gain information all the time, for example, on Tuesday when NRL team lists are announced, on Thursday when it is announced that 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 issues a weather alert for a severe thunderstorm. All these factors impact the possible outcomes of the match and therefore impact the models. Stats Insider's suggested investments may change frequently right up until game time.
How do Stats Insider's in-game simulations work?
Our live models work in a very similar way to the pre-match ones. However, 10,000 simulations take a fair bit of time and computing power to run, and this cannot be done every minute during the game. Depending on the sport, 500 or 1,000 simulations of the rest of the game may be run. The approach is the same: if the rest of the game were played a certain number of times, how many times does each team win?
How do Stats Insider's futures tips work?
Our seasonal projections take two primary inputs: the individual match simulations, where each remaining match in the season is simulated 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 due to injury, or if they are in doubt for the coming week, the season simulations will capture this and force the team to 'play' some of those games without that player. As with the single event predictions, these probabilities are then lined up with the bookies' odds to determine if there is a bet to be made.
What should I do with the information?
Well, that's up to you. We place bets on many of our suggested investments each week. We also enter them into tipping contests and use the information to inform our fantasy decisions when picking our daily fantasy teams. Ultimately, it's your decision how to use the data we provide.