I was analyzing some trades and ruminated on some principles. Probably obvious to most but I find I can keep learning by going over stuff I think I "know".

I thought I would share:

1) You only make money in a trade if price moves in your direction after you enter. This is a simple concept but the process has so much depth.

2) We can abstract the market into the crowd, aka customers, and market makers (MM). Price is set by the interaction between the crowd and the MM. MM only get involved if 2 members in crowd can't agree on price.

3) So, one thing should be very clear: that data series that we are so good at recording, quantifying and "understanding" of PAST prices is only good in our FUTURE trading to the extent that the interactions of the crowd and the MM are confined to the same boundaries as the past.

4) And these boundaries are often not there. Laws change. Exchange rules change. The way information is disseminated changes. Fashions change. Etc. These changes are not easily observed in price data. Also, many future changes are not even estimable.

5) So, while a lot of time and money is spent looking at past prices, in my opinion what you really need to focus on is the crowd and the market makers. Position, flows, changes, etc.

Prices help you understand the thing but they aren't the thing.

6) If this is our approach then we should immediately understand that longer term predictions are going to be very, very hard. But it should also be clear that there will be instances that it will be quite easy to predict at least certain members of the crowd's actions at times.

7) Good examples are the players with known algos: anything passive (VFINX, XIV, BOIL, etc), corporations with stock buybacks that follow 10b-18, and central banks, etc. There are many large, algo following crowd members.

8) So, then the problem becomes: What is the rest of the crowd going to do? My shorthand term for this is "I need to frontrun the frontrunners" This is not easy to figure out and where a trader proves himself and where we have to work a lot..

9) WRT MM you must know that they change over time. Right now they are very "quant". You can read their books and papers. They tell you [almost] exactly how they behave.

10) But the most important thing to remember is this: if the spread is small they cannot take much risk/inventory. So, if flows become one way they will lift offers/bids. There is no way around this fact. And spreads are so very, very small these days.

11) So then our job as traders is to a) find markets where we are good at observing the crowd and predicting the interaction between the crowd and the MM and then b) finding the best implementation on that prediction.

12) A great little video about how price <> market is by a big bond MM/trader in Australia named Rambo:

13) For me one way I approach the above is I look for big carries and try to figure out when/how/signs of a blow up. And usually my preferred implementation is some type of long option. Long gamma + long vega is best implementation if I am right.

14) But there are obviously many, many good approaches. Have to fit your interest, personality and skill set.