TrendSlug

- Sharpe ratio - Imagine if the distribution of returns had a regular (Gaussian) bell curve shape. The Sharpe ratio shows how wide the bell is. Steeper bell is better. - Sortino ratio - Same as Sharpe, but we only care about the left half. Steep left half is better (smaller tails). - Omega ratio - We know that the distribution of returns is non-gaussian (the bell has irregular shape). Let's measure the the area under the curve for the left/down part and for the right/up part. The ratio between the two is called the omega ratio. - Rachev ratio - Same as Omega, but let's find only the degen assets, we only care about the tails of the curve. Let's find which ones have the fattest tails. en.wikipedia.org/wiki/Rachev_ra…

New year, new bull-market. Are we ready?

After the recent dump, the major #Crypto assets now have negative 1Y Sharpe ratio. On a more positive note, the predictions have now turned seriously bullish.

The situation seems ripe for a bounce. At least statistially... $BTC, $JLP

@FpremL No, I trade my gold via ETFs, but looks ok

After today's flush, for the first time in some weeks, I'm starting to see positive signals for the major cryptocurrencies $BTC, $ETH, $SOL. Also, notably $PAXG seems to want go up in contrast to the $GLD ETF. But... FWIW, $PAXG seems to be a better indicator for the markets than $GLD itself. That's mostly because it has perps, which are also used for the prediction.

OK, but how significant is this effect? Pretty significant. Out of all the 150 parameters that go into the learning, retrograde mercury in in the third place by effect. Comparable in magnitude to the momentum of #FED inflation expectations for the next 5 years.

OK, but does it mostly affect #crypto? Nope, it has even more pronounced effects on some #TradFi assets. In particular $NVDA seems to be one of the most affected by this. $BTC, $PEPE, $NVDA, $SIL, $GDX, $URNM

This started as a joke... but then it turned out to be real... People seem to be using astrology to trade and by that they influence the markets. Hence, astrology influences the markets. What we see here in the screenshot is the 14 day predictions issued by the algorithm yesterday and today. Can you see what's different? Yes, retrograde mercury starts in 14 days from today... So this parameter now has a negative effect on the prediction. $BTC #astrology

Yesterday's $BTC candle was massive. It noticeably changed the 1Y Rachev Ratio, this was the worst candle for the past year. Rachev ratio quantifies tail events, so in a way this was THE black swan of the year. We can interpret it both ways. Bullish - this was the worst candle of the year so we are pumping now. Or, Bearish - This was the worst candle of the year, therefore trends have changed bearish. en.wikipedia.org/wiki/Rachev_ra…

Implemented a prediction explanation page where predictions can be compared by date, asset and model feature. Soon we will allow more people on our platform. Follow for more.

Been working on adding macroeconomic data to the learning parameters. Things like BAA10Y, T10Y2Y, T10Y3M. The first one of these seems to have very good predictive power for the overall market direction. At least the ML algorithm finds strong statistical correlation between it and the market moves over the next weeks. Latest move is quite a heavy dump since the beginning of Sep and around the middle of the month it broke the support it's been trying to break since June. So far Uptober seems... plausible? fred.stlouisfed.org/series/BAA10Y

I don't know guys... Is it time for $FARTCOIN?

Over the past days I spent some time implementing the so called "risk adjusted measures" - Sharpe, Sortino and Omega ratios. Of the three, Omega turned out to be the "best" for what I am doing, so now it is the default color gradient in the chart (red to green). Here is a good article on Omega ratio: swanglobalinvestments.com/advisor/omega-… On top I implemented the so called "Pareto Frontier" (links in bottom) in three dimensions: trend score, prediction and omega ratio, so now they are displayed as upwards/downwards (for the downside Pareto front) in the chart and it helps noticing some interesting assets which might get lost among the others. So here is a comparison of some major assets (both crypto and TradFi) employing these principles Yes, $PUMP looks interesting for longing here... and yes $CRCL is a bit scary. Wonder what it tells us about the state of the crypto market. en.wikipedia.org/wiki/Pareto_fr… en.wikipedia.org/wiki/Maxima_of…

@0x_Kun @goodalexander "Betting on winners" is the definition of trend following.

Agree with sentiment but the eg you give is not indicative Top blasting previous investments is betting on winners established already usually already within his book - not about betting on startups in a already trend phase when your thing isn’t 10x better to overcome This is within his framework Nonetheless even with best intentions no one can fully transfer what they know as it’s limited to how they can articulate

I met with Peter Thiel once I pitched him on a system to detect trends he told me it wasn't interesting because by the time a trend is in place, it means that it's likely overvalued I thought I learned a lost then he top blasted all his previous investments and turbo printed this was probably the first time I internalized that powerful people have no real reason to tell you the truth behind their process. you can just say things. You can show up to Market Wizards and fabricate your entire process and nobody will know. all rhetoric is just a methodology to reach an objective, especially with the powerful, there is zero truth value applies to many areas of life TruthCels will read Warren Buffet's newsletter and buy value stocks and underperform for decades while Buffett full ports AAPL at 26x earnings and sells the highs everything you think you've learned from people is probably just some sht they said in the moment, either to give them a compliance benefit. cheapen a deal. meme a founder into doing something. who knows you certainly don't. I certainly don't. the only data worth anything is first party study / collection

And we can do this also for stocks, where (comparatively) $BTC seems to have the best prediction looking forward.

And yes, if we look at predictions (only) 42 days forward, we see mostly positive predictions:

For the very short term (2-7 days), the predictions are still (very) negative but... (a thread)