Streaks

[modagavr]

Streaks

[geekzsp]

My strategy is to monitor the price and stop betting when the price swings wildly。

[geekzsp]

Or maybe you got lucky on the first day, and how to predict the trend is a conundrum

[CRuzylo]

My strategy is to monitor the price and stop betting when the price swings wildly。

I understand. This is what seems to be best. However I am really looking for this to get ran 24/7 in the background. It seems if a simple option to turn on or off auto betting the same way, it would really be beneficial.

Especially today, where the downtrend of crypto stayed strong the past 24 hours.

After reviewing how everything ran,
There were many instances of 8-10 in a row in one direction. This change would have resulted in a majority of today having been wins for the bot rather than otherwise.

The bot definitely excels during a stabulized market. I think if we can solve its consistent betting against the trend, it would really shine & add a good 5-10% to the average win rate

[geekzsp]

You can code to realize your ideas, it shouldn’t be difficult CRuzylo ***@***.***>于2021年9月20日 周一下午3:42写道：

…

My strategy is to monitor the price and stop betting when the price swings wildly。 I understand. This is what seems to be best. However I am really looking for this to get ran 24/7 in the background. It seems if a simple option to turn on or off auto betting the same way, it would really be beneficial. Especially today, where the downtrend of crypto stayed strong the past 24 hours. After reviewing how everything ran, There were many instances of 8-10 in a row in one direction. This change would have resulted in a majority of today having been wins for the bot rather than otherwise. The bot definitely excels during a stabulized market. I think if we can solve its consistent betting against the trend, it would really shine & add a good 5-10% to the average win rate — You are receiving this because you commented. Reply to this email directly, view it on GitHub <#19 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AFS3T3GNJFWGYPN5STXTSZ3UC3QURANCNFSM5EKO5WGA> .

[eugenepro2]

My strategy is to monitor the price and stop betting when the price swings wildly。

I understand. This is what seems to be best. However I am really looking for this to get ran 24/7 in the background. It seems if a simple option to turn on or off auto betting the same way, it would really be beneficial.

Especially today, where the downtrend of crypto stayed strong the past 24 hours.

After reviewing how everything ran,
There were many instances of 8-10 in a row in one direction. This change would have resulted in a majority of today having been wins for the bot rather than otherwise.

The bot definitely excels during a stabulized market. I think if we can solve its consistent betting against the trend, it would really shine & add a good 5-10% to the average win rate

I can do turn on, turn off via telegram bot. What do you think about this?

[CRuzylo]

@modagavr May you look into this?

for example, predictions went down 9 times at one point today, if the bot triggered to follow the trend before the 4th bet starts, it would have won 66% of the time, guaranteed. When the trend breaks, have it bet as normal until a new trend forms.

[eugenepro2]

How to follow the trend?

[YveOms]

This is an analysis of a (Bull-Bull-Bull) trend I've done using Python. It's a graph of (Amount remaining in your wallet (in Kenyan shillings $1 ~ Ksh. 100) after a round ends less all deductions less average network fees) against (Rounds or "Epochs" as I've called them).

As you can see from the trends there are times there is a bullish trend up to 8 times in a row, and twice there was a bearish trend of up to 6 in a row.

Thanks to @modagavr for helping me achieve all this.

@CRuzylo Honestly, predicting that the next round to be played is going to for a trend is so difficult. From simulations trying to bet for trends results in a horizontal profit margin/ loss (after a while you either have + or - 5% of your initial investment).

[CRuzylo]

This is why I recommend it not to predict it, but to determine when a trend may begin, if you consider in your graph above that on all trends, it wins everything after the first 3 epochs in one direction, & only in this case would it trigger, then you would have had a much more positive win rate, even assuming the initial 3 on each trend is possible wrong, in a worst case scenario.

I do get what you mean, truly. But the bot preforms incredibly well when its more stable & absolutely gets demolished on trends. This in my eyes would alleviate the issue.

The hardest part is having the bot determine if a trend is setting by before the end of the third bet. While it may result in an inconsistency, the bot should compare the price of chainlink maybe 30 second before close to the lock price & make a guess on whether the 4th will continue the trend, based on the current difference of chainlink to the lockprice. It would then, assuming the 3rd won, push the 5th to continue the trend no matter what, where if the 5th lost & the trend breaks, then it would restart its process, however if the forth one, then the 6th bets up. If the 5th lost, the 7th doesn’t bet on the trend, but as normal.

i know that’s rather complex. But atm, I think this is one of the few ways to improve it struggling against trends.

maybe add in —exp2 or something and allow users to test if implemented in a separate strategy.

@modagavr @geekzsp @YveOms

[YveOms]

@CRuzylo Hit me up on my telegram account @kaka_oms. We can try to make this work.

[YveOms]

@CRuzylo It is happening now again.

[nnzo]

Hey @CRuzylo! What do you think about a private bot that I could possibly make for you based off of this one?

Obviously you won't need to pay upfront or anything, but super interested because of the amount you made on the first day 😂

Let me know :)