My (proper) addition to the thread:

Theoretically, if you can price a race perfectly, then you bet the overs and you win overall. Mathematically, that's a truism.

However, there are a lot of unknown's which will effect how "perfectly" you price a race. Here's what I often do: I price a race to 100% (something I always like to begin with, b/c it tells me their "true price" IMHO), then convert it to a varying percentage, based on an estimate of the level of unknown.

Lots of first up horse's in the race? Lots of queries. Maybe will price a value market to 60%. Full, exposed form, runners with reliable ratings? 80-90%.

It's basically an insurance policy when the unknown (which is what stops your market being perfect) becomes bigger.

As to the odds on thing, I stay out of odds on runners, and will look for value around them, even if I price them odds on.

Here's a good example - last Saturday I priced (at 100%) Candy Vale $1.80 and Demerger $6.50. Next best, $12+. I thought Candy Vale was a special, I rarely price them in the odds on, esp in Saturday metro's.

However, I had one bet in the race. Demerger. For obvious reasons.

Had I priced Candy Vale $1.60 and $1.80 or $1.90 (or probably even $2.20) was available, I'd still have balked. Why? Because there's one BIG thing that happens to odds onnies that doesn't happen to 10/1 shots. They run them dead and bet around them.

Candy Vale wasn't dead. She was dead unlucky but she wasn't dead. But a SIGNIFICANT enough number are run dead to ensure that if you take odds on, even if you price it overs, you'll get skun.