So, I guess the bosses have just remembered that I'm working down here. I'm not sure if they realise that I've actually moved in. I was spending so much time here tinkering with the droid that eventually I just stopped going home. That's dedication, if you ask me, but apparently they need me to be writing articles and coming up with tips, otherwise there's no point having me here.

Well, I guess there is a lot to report. You'll remember that I was working on getting the Tipdroid right, and then I was calling it the Laydroid because it was doing such a great job of picking losers. Then I designed two separate modes for it, Laydroid mode and Placedroid mode. The Laydroid was a great success, the Placedroid not so much and in my tinkering to fix the Placedroid, I may have accidentally ruined the Laydroid. I lost confidence in it and I suppose I've been a bit quiet ever since. But I've been very busy working on the next incarnation and I think I have something that can be very useful, particularly if you understand its limitations.

Allow me to introduce The Quadinator.

Punting Form's System Builder is a remarkable tool that can help punters make informed bets with only a few minutes spent on the form guide, and I realised that with unlimited (and free) access to not only the builder but also Punting Form's unique sectional data, I could do things that the average punter could not. While most users of the System Builder will only subscribe to a limited number of systems (as they will have to pay for each subscription), I can build and subscribe to as many as I want. This allows me to create dozens of "sub-systems" - systems that may not be profitable on their own, but can highlight runners that stand out from the field in some way.

So, this is how The Quadinator works. Every morning, I spend about 10 minutes transferring information from the Punting Form database into the machine. It then sorts runners by race and presents me with the horses in each field which have triggered a hit on a profitable system or any of the sub-systems. Most races will have at least one runner that stands out from the field in some way and most races will have 2-4 such horses. Generally, the higher the quality of the race, the more runners that will get at least one hit and strong fields could have as many as 6 or 7.

The next step is to feed Punting Form's sectional ratings to the droid. With this data, it can identify further contenders and/or strengthen the case for the system and sub-system hits. By this time there will usually be 1-3 runners per race marked as strong contenders with possibly several more that have attracted The Quadinator's interest in some way. In 10-15 minutes, The Quadinator can tell a story of an entire meeting, yielding a shortlist for every race that might have taken hours to compile the old-fashioned way.

The question then became how to best use this information. At first, I just tried to pick out the individual horses that hit the most strongly and bet on those. It was marginally profitable but in too many races, my pick was beaten by one of the other runners that The Quadinator had put forward. Race after race was won by a horse on the droid's shortlist and rather than try to predict which of them would be the winner, I changed direction. I dabbled in trifectas and first fours, but the main focus became quadrellas and it was then that the droid was given its new name.

I've been using it for about a month now, putting on a quaddie at every meeting using the runners from The Quadinator's shortlists. It usually results in somewhere between 60 and 200 combinations, and over the month it's picked close to a dozen quaddies and I've won back slightly more than I've spent. With quaddies, I'd need a much, much bigger sample size before I could say it's a profitable method, but I'm optimistic.

Still, there are some things that no machine can do better than a knowledgeable punter. An expert eye could go over The Quadinator's selections and sack a few useless ones. They could go over the form and find something that the machine missed. The average meeting for The Quadinator results in winners for three out of four legs, which means most days, there's one, maybe two winners missing. There are also a few useless selections that are making the bet more expensive than it needs to be. For that reason, I reckon it's best to use The Quadinator as a base and then take at least a quick look at the form for yourself, adding and subtracting as you see fit.

But if you don't have the time for all that, The Quadinator will provide you with informed selections in minutes. So, to keep the bosses happy, I'll be posting The Quadinator's selections a few days each week. You lot can then decide if/how you'll use that information and if you figure it all out, let me know.

Comments