for visibility. the sound gets better 1min in. GME rose 30% overnight thursday on no news - the media released some bullshit NFT marketplace release. The next day it tanked again and no press release from GME. media backpedaled and says NFT market has flopped, based on literally nothing.
make it make sense.
for visibility. the sound gets better 1min in. GME rose 30% overnight thursday on no news - the media released some bullshit NFT marketplace release. The next day it tanked again and no press release from GME. media backpedaled and says NFT market has flopped, based on literally nothing.
make it make sense.
If you aren't trading based on how the algos trade you are doing it wrong.Yeah I mean what do you expect? You act like this never happens. “Reportedly” aka rumor that they will do this. Media ran with it because of course they did, anything GME generates clicks these days. GME has said nothing of course, most companies ignore rumor shit like this all the time.
GME is a meme stock of course it ran on this news. This is a tale as old as time.
If you aren't trading based on how the algos trade you are doing it wrong.
reader's digest version:I see you mentioning this a lot, but what does it mean? Is there an algo algo somewhere we can scrutinize?
I wonder, is there a compendium somewhere of how algo's move. Last year, someone sent me a book link, but I was too new to get it.reader's digest version:
Algo trading is based on formulas and code. Therefore it needs data points. For example, all those moving averages we talk about. For another, news/social media headlines and posts.
There is nothing "fundamental" that makes a stock price $5 above the 100 DMA different than the stock price at the 100 DMA. But, its a data point that algos are programmed to move on. No one would give a shit about technicals if algos weren't programmed to move massive amounts of money on those technical data points (among God knows how many others).
So, I don't buy a stock on the 50/100/200 DMA because it means anything rational. I buy it because I know the algos are programmed to act on it and they move a metric fuckton of volume which in effect becomes a self-fulfilling prophecy. Enough algos get programmed to buy at the 50-DMA on the SPY and the QQQ that so much money gets pushed into buy orders by them that it ends up moving the stock upwards. Now add in everyone who does the exact same thing (buy off the 50-DMA because they know the algos will do it) and you have created "support" levels out of thin air.
Where do you buy on this 1-year chart of the SPY?
View attachment 392136
Now where would you buy on this 1-year chart of the SPY?
View attachment 392138
If you bought on/around the 50-DMA, 9 out of 10 times you made the right call.
They are all individually written and very proprietary. I would be curious how someone could track their performance.I wonder, is there a compendium somewhere of how algo's move. Last year, someone sent me a book link, but I was too new to get it.
Dont forget the premium fiber laid across oceans so that your algo gets news faster than the other algos.They are all individually written and very proprietary. I would be curious how someone could track their performance.
But they behave similarly even if the implementations and optimizations are different.They are all individually written and very proprietary. I would be curious how someone could track their performance.
Nah man they use microwaves. Didn’t you read Flash Boys?Dont forget the premium fiber laid across oceans so that your algo gets news faster than the other algos.
reader's digest version:
Algo trading is based on formulas and code. Therefore it needs data points. For example, all those moving averages we talk about. For another, news/social media headlines and posts.
There is nothing "fundamental" that makes a stock price $5 above the 100 DMA different than the stock price at the 100 DMA. But, its a data point that algos are programmed to move on. No one would give a shit about technicals if algos weren't programmed to move massive amounts of money on those technical data points (among God knows how many others).
So, I don't buy a stock on the 50/100/200 DMA because it means anything rational. I buy it because I know the algos are programmed to act on it and they move a metric fuckton of volume which in effect becomes a self-fulfilling prophecy. Enough algos get programmed to buy at the 50-DMA on the SPY and the QQQ that so much money gets pushed into buy orders by them that it ends up moving the stock upwards. Now add in everyone who does the exact same thing (buy off the 50-DMA because they know the algos will do it) and you have created "support" levels out of thin air.
Where do you buy on this 1-year chart of the SPY?
View attachment 392136
Now where would you buy on this 1-year chart of the SPY?
View attachment 392138
If you bought on/around the 50-DMA, 9 out of 10 times you made the right call.
As I recall, not only did investors lay their own wire from the West Coast to Wall Street (that was shorter than all the other commercial connections) they also built their trading servers right next to Wall Streets servers. Like physically in the same room.Nah man they use microwaves. Didn’t you read Flash Boys?
I think you said this much more eloquently before. Something like...reader's digest version:
Algo trading is based on formulas and code. Therefore it needs data points. For example, all those moving averages we talk about. For another, news/social media headlines and posts.
There is nothing "fundamental" that makes a stock price $5 above the 100 DMA different than the stock price at the 100 DMA. But, its a data point that algos are programmed to move on. No one would give a shit about technicals if algos weren't programmed to move massive amounts of money on those technical data points (among God knows how many others).
So, I don't buy a stock on the 50/100/200 DMA because it means anything rational. I buy it because I know the algos are programmed to act on it and they move a metric fuckton of volume which in effect becomes a self-fulfilling prophecy. Enough algos get programmed to buy at the 50-DMA on the SPY and the QQQ that so much money gets pushed into buy orders by them that it ends up moving the stock upwards. Now add in everyone who does the exact same thing (buy off the 50-DMA because they know the algos will do it) and you have created "support" levels out of thin air.
Where do you buy on this 1-year chart of the SPY?
View attachment 392136
Now where would you buy on this 1-year chart of the SPY?
View attachment 392138
If you bought on/around the 50-DMA, 9 out of 10 times you made the right call.
As a previous telecoms engineer, it wasnt really about the fact they built their own as much as it was the optical distance shortening. Normal high speed fiber lines still snake and weave around natural and manmade obstacles. The Flash Boys effort was all about a "at any cost" direct linking with as few optical bends in the cabling as possible. Quite literally direct lines through mountains! Overall for this effort, shaving a mere milliseconds off of data exchange (an eternity in the algo world). In high speed algo trading, you're essentially cutting to the head of the line between say the Chicago exchange to the NY exchange and gaming the price spread with your edge on the NY exchange before that order actually arrives over other conventional transmission lines.As I recall, not only did investors lay their own wire from the West Coast to Wall Street (that was shorter than all the other commercial connections) they also built their trading servers right next to Wall Streets servers. Like physically in the same room.
Nobody realized this was an issue for like 14 years.