Overview
High Level Idea
Our miners act like traders. To score well and receive incentive, they place orders on our system against different trade pairs. The magnitude of each order is determined by its leverage, which can be thought of as the percentage of the portfolio used for the transaction. An order with a leverage of 1.0x indicates that the miner is betting against their entire portfolio value.
The first time a miner places an order on a trade pair, they will open a position against it. The leverage and directionality of this position determines the miner’s expectation of the trade pair’s future movement. As long as this position is open, the miner is communicating an expectation of continued trade pair movement in this direction. There are two types of positions: LONG and SHORT.
A long position is a bet that the trade pair will increase, while a short position is a bet that the trade pair will decrease. Even if the overall position is LONG, a miner can submit a number of orders within this position to manage their risk exposure by adjusting the leverage. SHORT orders on a long position will reduce the overall leverage of the position, reducing the miner’s exposure to the trade pair. LONG orders on a long position will increase the overall leverage of the position, increasing the miner’s exposure to the trade pair.
Basic Rules
- Your miner will start in the challenge period upon entry. This 90-day period will require your miner to demonstrate consistent performance, after which they will be released from the challenge period, which may happen before 90 days has expired. During this period, they will receive a small amount of TAO that will help them avoid getting deregistered. The minimum requirements to pass the challenge period:
- Score at or above the 75th percentile relative to miners in the main competition. The details may be found here.
- 10% Max Drawdown
- A miner can have a maximum of 1 open position per trade pair. No limit on the number of closed positions.
- A miner’s order will be ignored if placing a trade outside of market hours.
- A miner’s order will be ignored if they are rate limited (maliciously sending too many requests)
- There is a 10-second cooldown period between orders, during which the miner cannot place another order.
Scoring Details
Risk-Adjusted Returns is a heavily weighted scoring mechanism in our system. We calculate this metric using a combination of average daily returns and a drawdown term, assessed over different lookback periods. This approach allows us to measure each miner’s returns while factoring in the level of risk undertaken to achieve those returns.
While returns is a significant scoring mechanic, consistency and statistical confidence each play a substantial role in scoring our miners as we look to prioritize miners with a consistent track record of success. Additionally, we have a layer of costs and penalties baked into PTN, to simulate the real costs of trading.
We calculate daily returns for all positions and the entire portfolio, spanning from 12:00 AM UTC to 12:00 AM UTC the following day. However, if a trading day is still ongoing, we still monitor real-time performance and risks.
This daily calculation and evaluation framework closely aligns with real-world financial practices, enabling accurate, consistent, and meaningful performance measurement and comparison across strategies. This remains effective even for strategies trading different asset classes at different trading frequencies. This approach can also enhance the precision of volatility measurement for strategies.
Annualization is used for the Sharpe ratio, Sortino ratio, and risk adjusted return with either volatility or returns being annualized to better evaluate the long-term value of strategies and standardize our metrics. Volatility is the standard deviation of returns and is a key factor in the Sharpe and Sortino calculations.
Additionally, normalization with annual risk-free rate of T-bills further standardizes our metrics and allows us to measure miner performance on a more consistent basis.
Scoring Metrics
We use six scoring metrics to evaluate miners based on daily returns: Long Term Risk Adjusted Returns, Short Term Risk Adjusted Returns, Sharpe, Omega, Sortino, and Statistical Confidence.
The miner risk used in the risk adjusted returns is the miner’s average portfolio drawdown, the average of the maximum drops in value seen while we have been tracking the behavior of the miner. Once the drawdown surpasses 5%, it is no longer used directly as the denominator; instead, it is multiplied by a larger factor to amplify its effect. This emphasizes that a miner in the range between 5% and 10% average drawdown is riskier than a miner below 5% average drawdown.
To find the risk adjusted return, we take the annualized daily returns as the current miner return. We then divide this by the drawdown term. If, for example, a miner has a total 90-day return of 7.5% and a mean drawdown of 2.5%, their long term risk adjusted return would be 3.0.
Long term returns will look at daily returns in the prior 90 days and is normalized by the drawdown term.
Short term returns will look at daily returns in the prior 7 days. Besides the shorter lookback window, this calculation is the same as long term returns.
The sharpe ratio will look at the annualized excess return, returns normalized with the risk-free rate, divided by the annualized volatility which is the standard deviation of the returns. To avoid gaming on the bottom, a minimum value of 1% is used for the volatility.
The omega ratio is a measure of the winning days versus the losing days. The numerator is the sum of the positive daily log returns while the denominator is the product of the negative daily log returns. It serves as a useful proxy for the risk to reward ratio the miner is willing to take with each day. Like the Sharpe ratio, we will use a minimum value of 1% for the denominator.
The sortino ratio is similar to the Sharpe ratio except that the denominator, the annualized volatility, is calculated using only negative daily returns (i.e., losing days).
Statistical Confidence uses a t-statistic to measure how similar the daily distribution of returns is to a normal distribution with zero mean. Low similarity means higher confidence that a miner’s strategy is statistically different from a random distribution.
| Metric | Scoring Weight |
|---|---|
| Long Term Risk Adjusted Returns | 20% |
| Short Term Realized Returns | 10% |
| Sharpe Ratio | 17.5% |
| Omega Ratio | 17.5% |
| Sortino Ratio | 17.5% |
| Statistical Confidence | 17.5% |
Scoring Penalties
There are two primary penalties in place for each miner:
-
Max Drawdown: PTN penalizes miners whose maximum drawdown over the past 5 days exceeds the predefined 10% limit.
-
Martingale: Miners are penalized for having positions that resemble a martingale strategy. Two or more orders that increase leverage beyond the maximum leverage already seen while a position has unrealized loss may result in a penalty. More details on this can be found here.
The Max Drawdown penalty and Martingale penalty help us detect the absolute and relative risks of a miner’s trading strategy in real time.
Fees and Transaction Costs
We want to simulate real costs of trading for our miners, to make signals from PTN more valuable outside our platform. To do this, we have incorporated two primary costs: Transaction Fees and Cost of Carry.
Transaction fees are proportional to the leverage used. The higher the leverage, the higher the transaction fee. We use cumulative leverage to determine the transaction fee, so any order placed on a position will increase the fees proportional to the change in leverage.
Cost of carry is reflective of real exchanges, and how they manage the cost of holding a position overnight. This rate changes depending on the asset class, the logic of which may be found in our proposal 4.
Implementation Details
| Market | Fee Period | Times | Rates Applied | Triple Wednesday |
|---|---|---|---|---|
| Forex | 24h | 21:00 UTC | Mon-Fri | ✓ |
| Crypto | 8h | 04:00, 12:00, 20:00 UTC | Daily (Mon-Sun) | |
| Equities | 24h | 21:00 UTC | Mon-Fri | ✓ |
The magnitude of the fees will reflect the following distribution:
| Market | Base Rate (Annual) | Daily Rate Calculation |
|---|---|---|
| Forex | 3% | 0.008% * Max Seen Leverage |
| Crypto | 10.95% | 0.03% * Max Seen Leverage |
| Equities | 5.25% | 0.014% * Max Seen Leverage |
Leverage Limits
We also set limits on leverage usage, to ensure that the network has a level of risk protection and mitigation of naive strategies. The positional leverage limits are as follows:
| Market | Leverage Limit |
|---|---|
| Forex | 0.1x - 5x |
| Crypto | 0.01x - 0.5x |
| Equities | 0.1x - 5x |
We also implement a portfolio level leverage limit, which is the sum of all the leverages from each open position. This limit is set at 10x a “typical” position, where a typical position would be 1x leverage for forex, 2x for equities, and 0.1x leverage for crypto. You can therefore open 10 forex positions at 1x leverage each, 5 equities positions at 2x leverage each, 5 forex positions at 2x leverage each, 5 forex positions at 1x and 5 crypto positions at 0.1x, etc.
Incentive Distribution
The miners are scored in each of the categories above based on their prior positions over the lookback period. Penalties are then applied to these scores, and the miners are ranked based on their total score. Percentiles are determined for each category, with the miner’s overall score being reduced by the full scoring weight if they are the worst in a category.
For example, if a miner is last place in the long term realized returns category, they will receive a 0% score for this category. This will effectively reduce their score to 0, and they will be prioritized during the next round of deregistration.
We distribute using a softmax function, with a target of the top 40% of miners receiving 90% of emissions. The softmax function dynamically adjusts to the scores of miners, distributing more incentive to relatively high-performing miners.