Each step towards speed moves the edge from the signal to the infrastructure
These are trading shorthands, not regulatory categories. HFT, high frequency trading, holds positions for very short periods, and its edge depends on reacting faster than others. MFT, medium frequency, holds for minutes to hours. LFT, low frequency, holds for days, weeks or months. Each label describes the time scale of a strategy, not a licence.
The chart puts the three on one path. The further towards speed a strategy sits, the more its result depends on latency, the order path and hardware. The further towards holding, the more it depends on the signal, the costs and the risk carried overnight.
Speed is bought in layers: a faster network link, a server placed close to the exchange, lean code, and a risk system that can stop an order in time. Those layers cost money, which is why the fastest end is run by well resourced firms, and why the retail API route, with its order rate limits, is built for a different time scale.
A common error is to call any algo HFT. A program that places a few orders a day through a broker's API is not HFT, whatever its label. The frequency label should match what the system does, as shown in its own logs. The question on why traders lose covers the costs that a fast strategy must also beat.
Before you trade. Derivatives trading carries a high risk of loss. SEBI's study "Profitability of Individual Traders in the Equity Derivatives Segment (FY25 to FY26)" (August 2026) found that 87.7% of individual traders in equity F&O made net losses in FY26, with aggregate net losses of Rs 91,685 crore. Read the SEBI study