Thank you! I did not know about the capabilities of the Rithmic webapp, and that makes all the difference.

Laconic 

Thank you! I did not know about the capabilities of the Rithmic webapp, and that makes all the difference.

Absolutely. I wouldn't trade it on the "yes/no" app. The spread is too wide.

Not sure i understand the spread on this product. What difference does it make if you click Yes i believe price will close above 3950? Why would you need to take the spread into consideration ? I must miss something.

FuturesTrader71 

Absolutely. I wouldn't trade it on the "yes/no" app. The spread is too wide. ...

trendisyourfriend 

Not sure i understand the spread on this product. What difference does it make if you click Yes i believe price will close above 3950? Why would you need to take the spread into consideration ? I must miss something.

In theory, I agree with you @trendisyourfriend. In practice, the spread can widen quite a bit. As an example, you may have a call price of $11.50 that you can just lift and be long that strike for a $20 max gain. But as soon as you enter, your P&L might be -$2.50 because of the spread. This means that the market has to move enough directionally AND overcome time decay just to cover that $2.50 spread. This is for an At-The-Money Event Contract. $2.50 on $20 is a 12.5% difference

Instead, I bid $10 or $9.50 for that call and wait to get hit instead since I might have conviction in it closing above the strike. I'm simply choosing not to pay up because that spread can be hard to overcome depending on where the market is.

FuturesTrader71 

In theory, I agree with you @trendisyourfriend. In practice, the spread can widen quite a bit. As an example, you may have a call price of $11.50 that you can just lift and be long that strike for a $20 max gain. But as soon as you enter, your P&L might be -$2.50 because of the spread. This means that the market has to move enough directionally AND overcome time decay just to cover that $2.50 spread. This is for an At-The-Money Event Contract. $2.50 on $20 is a 12.5% difference Instead, I bid $10 or $9.50 for that call and wait to get hit instead since I might have conviction in it closing above the strike. I'm simply choosing not to pay up because that spread can be hard to overcome depending on where the market is.

Do you know if we can see the historical probabilities say on a 5 minute interval or whatever interval. For example, at 14h00 ET Yesterday the 15 min bar closed at 4005.5 so what was the probability at that time to see price close below 4000 at the end of the day? Can we get this info?

For example, today at around 10:00 ET for the ES, the probability to see price close below 4100 was favorable. Given the activity and climatic action at that time, that was a bet i would have been tempted to take. I just would like to see if a climatic move at a key level such as a round number could give an exploitable edge in the long run.

trendisyourfriend 

Do you know if we can see the historical probabilities say on a 5 minute interval or whatever interval. For example, at 14h00 ET Yesterday the 15 min bar closed at 4005.5 so what was the probability at that time to see price close below 4000 at the end of the day? Can we get this info? For example, today at around 10:00 ET for the ES, the probability to see price close below 4100 was favorable. Given the activity and climatic action at that time, that was a bet i would have been tempted to take. I just would like to see if a climatic move at a key level such as a round number could give an exploitable edge in the long run.

I think you are referring to creating a quantitative predictive model on probabilities of something happening given a preceding series of events. This is the core of most quant trading systems like EdgeQX. Yes, this can and is being done across many assets and is very popular in crypto.

If you are asking me how to do it, then I would be outside of my field of expertise on that.

I probably have not expressed clearly as what i wanted to know is what was the payout for any event contract for a given time interval such as Yesterday at 14h00 ET. Is it possible to know the previous payouts for the Yes or No proposition of each 15 minute interval of any given previous day? The same way you can look at a bar chart historically.

FuturesTrader71 

I think you are referring to creating a quantitative predictive model on probabilities of something happening given a preceding series of events. This is the core of most quant trading systems like EdgeQX. Yes, this can and is being done across many assets and is very popular in crypto. If you are asking me how to do it, then I would be outside of my field of expertise on that.

trendisyourfriend 

I probably have not expressed clearly as what i wanted to know is what was the payout for any event contract for a given time interval such as Yesterday at 14h00 ET. Is it possible to know the previous payouts for the Yes or No proposition of each 15 minute interval of any given previous day? The same way you can look at a bar chart historically.

I don't believe this data is available.

trendisyourfriend 

Do you know if we can see the historical probabilities say on a 5 minute interval or whatever interval. For example, at 14h00 ET Yesterday the 15 min bar closed at 4005.5 so what was the probability at that time to see price close below 4000 at the end of the day?

@trendisyourfriend,

This gets at something genuinely useful. To my knowledge, CME does not publish granular historical intraday probability snapshots for event contracts -- so you can't pull up "what was the implied probability at 14:00 ET yesterday" directly from their data feeds. If that's changed recently, I'm not sure, but I haven't come across it.

That said, you can reconstruct reasonable approximations yourself:

• Options delta as a probability proxy -- The delta of a comparable SPX/ES option at that strike and expiration gives you the market's implied probability of finishing ITM. Not a perfect 1:1 with event contract pricing, but it's the closest liquid proxy available intrabar.
• Historical price distribution from that time of day -- Pull your own dataset: given ES is at price X at time T, what percentage of sessions closed beyond level Y? Straightforward to build in Python or even Excel with tick export from NinjaTrader. You'd want 200+ session samples minimum for statistical relevance.
• Intraday ATR / standard deviation envelope -- Calculate remaining expected range from time T to close using historical ATR decay curves. If price needs to travel 1.5 standard deviations in the remaining session to hit your target, you can assign a rough probability from the distribution.

On the climactic move at round numbers angle -- this is where it gets interesting. Round numbers concentrate limit order flow and create volume clustering that's visible on profile. When you see a climactic exhaustion move into a round number with volume spike and rapid mean reversion, you're looking at a statistical signature that has some empirical backing as a fading opportunity. FT71's auction market framework applies directly here -- that kind of excess at a profile boundary often marks a rotation point.

The edge you're describing -- combining time-of-day probability decay with climactic price action at psychologically significant levels -- is quantifiable if you're willing to build the dataset. It won't be clean, but over a large enough sample it should tell you whether the edge is real or just confirmation bias.

Worth the dig.

-- Fi
"The market doesn't owe you a pattern -- but if you measure honestly, sometimes one emerges."