Where Did the Money Actually Die: Flemington ROI -54.7% (Oct carnival)

My carnival ledger since 2 August says Flemington is where punters money goes to die, with ROI -54.7% from 27 bets.

Alright, let's answer the question everyone whispers to themselves right before they donate to the track: where has the money actually been made at the carnival tracks, and where did it go to die?

I've only got one kind of evidence that matters to me: my own public bets since 2 August. No vibes. No "this horse loves the track" bullshit. Just what happened.

The scoreboard: where it's worked, where it's hurt

Caulfield: the only place that feels vaguely civilized

At Caulfield, I've backed 56 times for 19 winners, striking at 33.9%. I staked $613.22 and the result was P&L -$67.54, which is ROI -11.0%.

So yeah, it's still losing overall. But Caulfield is the only carnival stop that doesn't feel like the horses are personally stealing my wallet.

Flemington: the money pit (27 bets, ROI -54.7%)

Flemington is the crime scene. I've had 27 bets there for 4 winners (14.8% strike). Staked $301.00 and P&L -$164.77, which lands at ROI -54.7%.

That's not "oh maybe you just need a bounce." That's "this pattern is big enough to at least demand respect," even if we should still be honest about sample size.

Randwick: winning strike rate, still losing ROI

Randwick looks better on paper and worse in the wallet. I've had 67 bets, 27 winners (40.3% strike). Staked $756.00 and P&L -$184.13, ROI -24.4%.

So the horses I've tipped have been winning at a pretty solid clip. The price has just not done me any favours. When strike is strong but ROI is ugly, that usually means the market is charging me for the privilege of being correct.

What we can say (and what we can't) without pretending

Two of these reads come with warnings.

• Caulfield's ROI is tagged as "more than luck?" as -11.0% with n=56, likely between -49.6% and +32.1%.
• Randwick's ROI is tagged as "more than luck?" as -24.4% with n=67, likely between -54.6% and +8.3%.
• Flemington's read is "early" at -54.7% with n=27, likely between -97.8% and -1.9%.

Translation: we can confidently say Caulfield, Flemington, and Randwick have all been tough for me. What we can't do is declare a permanent rule like "Flemington is cursed forever" and go full superstition.

Still, the direction is consistent enough to plan around it.

The extra punch: my "number-one picks" aren't saving me

Because nothing makes you feel like you're being pranked by the universe like your top confidence selections still can't drag the results over the line.

• At Caulfield, number-one picks only: 48 bets, 15 winners (31.2% strike). Staked $574.72, P&L -$71.44, ROI -12.4%.
• At Flemington, number-one picks only: 19 bets, 3 winners (15.8% strike). Staked $252.50, P&L -$122.57, ROI -48.5%.
• At Randwick, number-one picks only: 50 bets, 15 winners (30.0% strike). Staked $669.00, P&L -$204.19, ROI -30.5%.

So this isn't a case of "I got creative and paid for it." Even the bets I rate most have been getting rolled.

So what do you do with this before Caulfield Cup Day / Everest / Cox Plate?

You use it like a planner, not a tip sheet.

1) Treat Flemington as your "manage risk" track

With ROI -54.7% from 27 bets, Flemington is screaming that your prices and/or your angles are not aligned with what wins there for you.

Practical move: if you're personally strong at Flemington, keep playing your style. If you're not, tighten up. Less volume, smaller stakes, and be picky with the races where you're actually getting value, not just betting hope.

2) Randwick isn't a free pass just because strike is high

Randwick's strike rate is 40.3% across 67 bets, yet ROI is -24.4%.

That's the key lesson: winning more often doesn't automatically mean beating the market. If you're paying "correct price" money for "correct horse," you can still lose the war.

Practical move: focus on whether your selections are getting outpriced or just going where the market always intended. In other words, be suspicious when the market agrees with you quickly.

3) Caulfield is your "don't overreact" track

Caulfield has ROI -11.0% from 56 bets and number-one ROI -12.4% from 48 bets. That's still losing, but it's the least terrible of the three.

Practical move: Caulfield is where you can size closer to normal without immediately panicking. But don't get arrogant. A bad carnival doesn't need a dramatic villain. It only needs you to keep buying losing tickets at familiar prices.

The one thing I'd do differently at the big days

Since the whole point of a carnival is that crowds get loud and markets get slippery, I'd anchor your betting process to this question:

"Am I backing winners, or am I backing prices that let me keep my money?"

My ledger says: I've been backing winners often enough. The prices haven't been kind, especially at Flemington and Randwick. Caulfield is less brutal, but it still hasn't turned the taps back on.

If you want a simple plan for the upcoming big days, it's this: be cautious with Flemington bets, be value-focused with Randwick, and treat Caulfield as tolerable rather than "back the truck."

If you want to turn this into a checklist for Race 1s at Caulfield Cup Day, tell me what your usual approach is (early speed? ratings? vibes and binoculars?) and I'll translate the same data into a method you can actually follow without exploding your bankroll.