Stephen is one half of the duo behind our very successful AFL betting model. He’s on the podcast to run through each of the top 6 teams and give his Grand Final prediction.
He also talks about the changes in game styles, stats and the finals format this season.
Dave: Good day Stephen. Welcome to the show again.
Stephen: Yeah, thanks Dave. Good to be on.
Dave: Yeah. It’s been a really good season so far. I wanted to get you on and have a chat about how you see things panning out for the rest of the year. So we might as well start at the top and the Hawks have had a great run, but they’re not usually two games clear on top. How have they been, in terms of the numbers and your own projections?
Stephen: Yeah. I guess it’s pretty standard in the last few years to see them up top, but this year they’ve actually they haven’t played as convincing footy as might suggest from the table. By our numbers, we’ve projected them not quite as strong as in previous years. You’ve seen by the way they’ve won their matches this year, they’ve won several games by less than a goal margin against St. Kilda and Adelaide, for example.
It’s not the same dominance as we might have seen in previous years, but that being said, it is a sign of an experienced side getting it done in the close ones. I guess we can’t really hold anything against them at this stage. Now they’re in the box seat, two games clear as you said, to finish top two and have a home final in the first round in the finals series and then have potentially have a home prelim as well. They could be all games in the finals and MCG for the Hawks. It could be four in a row, back to back.
Dave: A bit scary, because they did it the hard way last year going to the West a couple of times.
Dave: This year they may not be leaving the state. You’ve gone to a heavy player ratings model.
Do you know this year compared to last if they’re actually producing at a higher rate or is it maybe the competition dropped away a little or what do the numbers suggest?
Stephen: In terms of player ratings specifically, they’ve would have dropped off a lot from the rating that we gave them last year due to loss of key players. Obviously, Jarryd Roughead the biggest of those, but even losing key defenders like Brian Lake and a few of their mids at times this year have been out, so we haven’t been overly lucky in the last few years, per se with injuries, but I think this year though, they’re rather undermanned and they still managed to get it done.
We probably saw a few more up-and-coming players going through this year, a few rookies like Kaiden Brand and few of the second year, third year players, Little and Whitecross and whatnot that you wouldn’t really put them as a high calibre of Hawthorn usually are, but that’s what they do. They take these new blokes on and then in a few years, they’re ready to win premierships. They’ve done well with what they’ve had this year, I think.
Dave: What about GWS? Obviously pretty much a team for first round draft picks and on the up. Do the numbers in your own models suggest that they could be a legitimate threat for the premiership?
Stephen: Yes. The way we rate them, we’ve got GWS currently as having the second best side in the comp. Obviously, that’s just on player ratings. That doesn’t take into account where they’re playing. Playing in Melbourne gives them a slight disadvantage, but as you said, the volume of draft picks they’ve got over the last few years is starting to shine through and even with a few injuries this year, they’ve show how strong their bench is.
I think they’ve rolled through about eight players not in their starting 22 yle between two at the start of the year. They’re all used to a pretty high level. In previous years though, starting by some key injuries to defenders and ruckman and while they’re not immune, I think they’re better set this year with guys like Rory Lobb really ready to go into the ruck to replace Mummy , and can get a spell with Patton if need be.
They did look very promising last year until a few key defenders and Mummy went down and then it just all fell off the rails. That being said, Mumford did go down last week, we eventually see how serious that injury is and hope he’s all right to finish out the campaign.
Dave: What about that other team out of Sydney? How do you rate the Swans?
Stephen: I think, for us, Sydney’s been a bit of a surprise factor this year, because start of the year they did have a lot of first-year players getting the call up like George Hewett, and Callum Mills has been a star. A few other players.
It’s always hard to predict how first-year players are going to go, even if they are high draft picks. I think they’ve exceeded our expectations, let’s say. Obviously, making SCG a bit of a fortress, besides the few losses there recently. I think it just shows their defensive structure, ability to withstand pressure inside the defense of the SCG is one of the best of the comp. They showed us that this year withstand barrages and inside 50s and then swing a coast-to-coast and make them pay.
I think they’ve really benefited from tough footy they’ve been playing the last few years and even with a few gaps in their 22, they’ve obviously made the most of it and they’re looking all right for about four finish.
Dave: Then we’ve got the Cats. That’s an interesting one, because obviously with the recruits over the off-season, there’s plenty of expectation, plenty of hype. How have they gone on your model so far and also the projections going forward?
Stephen: By the numbers they have improved significantly on both offense and defence from last year. We’ve got them as the most improved team in defence statistics and one of the top four in offense. Obviously, getting players like Patrick Dangerfield can help out that pretty quickly. The bash brothers in Danger and Selwood have been tearing a few teams apart when they get the chance. I think their consistency has been a bit of an issue at times, especially losing to teams like St. Kilda and Carlton, it’s tough to lose those games and then go on and win premierships.
They have taken down flag favourites Hawthorn and Adelaide. On their day, obviously, they can beat anyone, but I think that kind of inconsistency is going to come to get them at the end of the year. I probably wouldn’t put them as finishing top four with the competition up there and then I think it would be hard for them to make it all the way if they play away in a prelim final.
Dave: When you’re dealing with a model that’s about quantitative analysis, how do you handle a team like Geelong where they can literally beat anyone, but they can pretty much lose to anyone as well? How do you turn that into a projection?
Stephen: It’s tough, because we have our best estimate of what’s going to happen and that’s what we have to go by. It’s not a perfect world, obviously. With increased volatility usually increases value to the punter. We just have to trust in our best estimates and make sure that we’re prudent and being conservative and not to take anything willy-nilly or not to give them too much of an advantage where it’s unwarranted, especially little things like playing at Skilled Stadium against Victorian teams and stuff like that. The player ratings mode wasn’t quite hitting the mark with Brisbane with a few losses in a row at one point in the season. Other than that, we just tried our best estimates and rely on the research we’ve done over the last few years and then go off that.
Dave: What about the Eagles? You were on them pretty early last year at a good price and they fell at the final hurdle. They haven’t been nearly as impressive this year, according to most people watching it. What do you say?
Stephen: Yeah. I can’t explain the drop-off there. Obviously, some big injuries in the last few months, especially guys like Nic Nat, but he was there at start and they still didn’t look the kind of same force they did the year before. It could have been one of the things that the rest of the comp worked him out, similar to teams worked out Port Adelaide after their big rise a few years ago, but it’s tough to say. They have definitely gone down, both on the offense and defence side of things by the stats, which is to be expected compared to where there were last year.
It’s been heralded this year that they’ve been bad in the road, but they have faced some of the four hardest teams away in Sydney, Hawthorn, Geelong, and Western Bulldogs, so it’s tough ask to go interstate and beat those kind of teams away. It’s a tough ask, but if you’re going to really challenge the top four, you’ve got to pick up or two of those kind of games and they lost all four. Puts them in a tough spot now to come from the back and make a crack at the finals again.
Dave: Then we’ve got the Crows and there were a lot of people that wrote them off when they lost Dangerfield. You decided to back him at $34 for the flag. That’s looking a decent bet right now. How are they traveling?
Stephen: Yeah. They’re actually still our favorites to win the flag, despite the loss to Geelong last week. There are certain cases where the bookmaker really panics. Obviously, you lose, arguably, their best player in Patrick Dangerfield and the assumption is that everything is going to go out the window, but the reality was that that was strengthening all across the park, getting new players like Mitch McGovern coming into fill out the forward line. In terms of average player rating on the ground, they’ve been the highest this year in our model. Even just rattling off the names of their forward line, you can tell that it’s the most potent in the comp with Tex, Josh Jenkins, Eddie Betts, Charlie Cameron, and now, Mitch McGovern as well. There’s so much firepower there, which is why they’ve scored the second highest points this year behind GWS only by a couple of goals.
We saw potential there. I think it’s become obvious the longer the season’s going on. Now, the last few weeks the public has finally started to come around to them. Obviously, being close to a pick against Geelong, playing at Geelong was a big indication of where they’ve come in terms the market’s perception of them.
As I said, we still have them as favourite to win the flag from here with a highly likely case that they finish in the top four as well. I think that would make a good crack at the post-season and, hopefully, for our sake and theirs, the injuries to Daniel Talia and Taylor Walker over the weekend aren’t too serious.
Dave: For the Grand Final, are you expecting Crows/GWS or Crows/Hawks?
Stephen: Swans are actually our second favourite at the moment.
Stephen: From the sims we’ve been running, we project GWS and Sydney to finish four and five, in which case, if that happens, there’d most likely be a Sydney qualification final. I’d say one of those two highly likely. It would take someone like … If Hawthorn finished first, it’s obviously going to take someone to beat them at the MCG, but I think a team like Adelaide, Sydney, or GWS could go out and do that.
From the model, most likely case is Adelaide vs Sydney, GF, which would be awesome for a Sydney boy to see two Victorian teams in there.
Dave: Does the new structure of the last week of the season change anything for you, because in the past the big advantage to finish top four and win was you get a week of, but now everyone gets a week off before the finals. If you get that extra week off, you’ve only got the one game in three weeks. Does that have any effect?
Stephen: I think, obviously, the purpose of it was to not make round 23 a joke kind of thing and make a full round of footy without coaches being tempted to rest. I think it will achieve that. I think that’s a good thing.
In terms of the rest and whether that’s a benefit or a disadvantage, historically from our research we’ve found there is evidence for say a one-week break for a slight advantage assigned to a team with a break versus a team that hasn’t had the break. That makes sense, because, if coaches are inclined to rest players, and obviously it’s for a greater good, in terms of two weeks off in three weeks, I think given the fact that the top four teams that win in the first week of finals, having a home prelim final is such a big advantage, especially for a team that is probably going to play an interstate team, that it will offset any kind of disadvantage that they may have from less footy than their opposition in that three week period.
There’s also the fact that every week is another chance for a key player or any player in the starting 22 to go down with an injury. It also gives that coach peace of mind, their best 22 after week one of the finals is going to be the same in the prelims. I think there’s a lot of benefits that way and the negatives that way.
Dave: Okay. Just overall for this season, you dig pretty deep into the stats. Have you noticed much of a shift from last season? There’s been rule changes and game style changes and some coaches have gone with different tactics. How has that been incorporated into what you do, which is trying to look at the past to predict the future?
Stephen: Yes, it’s tough, because real-time coaching and strategy changes really don’t flow through the model immediately. They proportionally enter them, if those strategies create a different kind of performance. Say the team’s stats and performance is improving from last year, it will proportionally start to take its effect in the data, week to week, more so.
We don’t really have any manual adjustments for what we see in game strategy. We kind of let the data take care of itself there. We’ve definitely seen a few patterns this year. Start of the year, a few rule changes, a few more free kicks being paid for various things. Totals, the points scored seems to be increasing. Then, the last few weeks, as winter has really hit, some cold night games, maybe even the introduction of the extra umpire has helped as well, but, yeah, it’s been a trend downwards in point scoring.
I think all in all for the season it hasn’t been too different from last year once you take the last few weeks into account. Obviously, it’s changed by team, like you’ve seen Hawthorn really hasn’t been the juggernaut on offense and goal kicking machine that we’ve known in the past and that’s more been taken up by GWS and Adelaide. Melbourne has really improved off the map, benefit of another year of being under Roos there and just a few more solid players coming into their own. Jack Viney and Bernie Vince and Christian Petracca and Max Gawn. I think in terms of how our model allows for it and deals with it, yeah we do proportionally take the data into account, but we don’t do anything based on strategy until it comes through.
Dave: There must just be anomalies at times, because we were on the Dees on the weekend that covered the line, no problem. They wee never not going to cover.
Dave: We’re on pretty close to $6 and they absolutely dominated inside 50s, but just couldn’t get the win. I think that’s happened a couple of times this year where teams have lost the entire 50s badly, but still won. Is that normal?
Stephen: I mean, there is a significant enough relationship between differential inside 50s and margin of victory historically, but, yeah, as you said, there’s anomalies in every game. There was two factors in the weekend that helped West Coasts’ chances there. One was that it was extremely wet, so each inside 50 isn’t really worth as much as it would be in a dry game. Chances of getting a mark inside 50 were a lot lower in those kind of conditions and that decreases as your scoring opportunities and your percentage of getting a goal from each inside 50.
Another thing with a lot of these inexperienced teams as far as what they do with the ball inside 50 is very different to what a more experienced team might do. You watch the Hawks or Sydney, if they’ve been withstanding to any pressure for a while, they get an opportunity to go inside they’ll lower their eyes and get a good shot 30, 35 metres out. Whereas a lot of these inexperienced teams will might just have the panic bomb in the middle of the forward 50 and just hope for the best, which on the weekend, obviously, didn’t work out and in those conditions is even worse a strategy.
I think it’s a combination, in our case, weather, and as well method or strategy of using the ball inside 50, which inexperienced teams seem to be worse at than the better ones.
Dave: For sure. All right. We’ll leave it there for now. Stephen, it’s been a hell of a year so far, so congratulations on that. We’ve got a lot more people now following these selections, yeah, no pressure at all for keeping it going for the rest of the season.
Stephen: Good to hear. Thanks for having us on, Dave.
Dave: No worries. Cheers.