All models are wrong, but some are useful.
George Box, Statistician
The year 2011 had been mildly down for MIDAS but a brutal one for the market benchmarks. Investors were quite pleased with the performance, this being a long only strategy. Ajay Kaul was also toying with an idea of a long/short absolute return fund if Midas-long only gave good results over a five year period in fairly liquid stocks. But work on that would have to start now.
In the meantime he ruminated over his debrief session with Rahul and the head trader, Chinmay. The key takeaways were:
- It was tough to surrender all decision making to a system and follow it slavishly. It took fair amount of detachment to keep one’s attention away from daily swings in price and news flow
- Human craving for narrative or stories is not fulfilled here.
- It was also very “boring” once the system was into production. And since it was not a high frequency machine, a part time “baby sitter” or analyst could check whether the out put of the system was matching the orders being placed. Also if any corporate action was coming up or there was any M&A play, those stocks would be flagged by the system. And other such checks and balances. It wouldn’t take more than half a day per week to run this system
- It was very tempting to override or tinker with certain parameters. In a back test , the system would take the entries and exits as programmed. But the human mind was a monkey. Primal fears still ruled our reptilian brain. It saw danger where there was none. It was still designed for survival in the jungle. It could not yet distinguish between actual danger and fears/danger created by the mind. This was precisely the edge that an investing system was supposed to capture.
“If you have a three year period where something doesn’t work, it ages you a decade.
Cliff Asness, Founder of AQR
Ajay also realised, that since running of the fund, meeting investors and running the discretionary strategy of the fund, kept him occupied and also more importantly kept him from interfering with day to day running of Midas. But those who were close to it, suffered swings along with performance of the strategy. Unlike the cool comfort of sitting in an ivory tower like an quant academic, a practitioner quant did not have the comfort of reworking his assumptions and updating his paper. They suffered actual losses which took a mental and physical toll.
However, 2012 did not pose that problem for Midas and Aks Capital. They had a good year in terms of absolute and relative returns. Midas (R) had it better though. Midas (R) had entered the year with only 2 longs and rest of the portfolio in cash. In the first two months of the year however, the mid cap index recovered, with the broader market, and Midas (R) deployed its cash reserves aggressively.
Midas was holding almost 12 positions by Feb/March 2013. It bought across industries like Pharma, FMCG, Cement and BFSI giving a fairly diversified portfolio in terms of sector exposure.
The year ended on a good note and it held some future multi baggers (with the benefit of hindsight) like Bajaj Finance, Eicher Motors, Kajaria Ceramics etc. The year’s big winners were Shree Cement (82%), and Kajaria (67%) which the fund still held. Most losers like Grindwell Norton (which it bought twice that year) and exited at relatively small losses of 10% each time. The simple mantra followed was winners stay, losers go. And there was no ego involved in buying back a loser, if the system ordered so.
So far, so good.
(With inputs from Aayushi Shah)
Disclaimer: Nothing in this blog should be construed as investment advice. This is purely for educational purposes only. Please consult an investment advisor before investing.