Tried them years ago and didn't like them is about the usual story, and the math backs it up. At an 85 percent payout, a win returns 0.85 and a loss costs 1, so the break-even win rate is 1 divided by 1.85, which is about 54.1 percent. That sounds close to a coin flip, but it needs the win rate sustained after any real delay, and it's the same on every trade no matter how clear the setup is.
Compare with my own best setup, a pullback entry that wins about 58 percent at 1:1. On a binary, 58 percent wins gives 0.58 times 0.85 minus 0.42, which is +0.073 per unit staked. Not terrible, but it ignores that the binary punishes me for a trade that goes 30 pips my way and then reverses by one tick at expiry. In spot, I'd have banked it with a target.
The structure is also what bothers me most, the platform being the other side of every bet. In a normal CFD account there's a conflict as well, but the payoff scales with the move, which seems fairer.
What did you try them on, short-expiry on majors? And what made you say they weren't suitable, the payouts or the way the trades felt?