QuantConnect/Lean · error · ValueError

Regression test failed: current ask price was not loaded fro

Error message

Regression test failed: current ask price was not loaded from NWSA backtest file and is not $1.1

What it means

This is a regression assertion in Lean's OptionRenameRegressionAlgorithm that verifies historical option quote data survived a corporate-action/symbol rename (the algorithm subscribes to TFCFA, which historically maps through NWSA, the old News Corp ticker). It checks that contract.ask_price for a specific CALL (strike 33, expiry 2013-08-17) equals exactly 1.1, proving the rename-mapped factor file and the backtest data file loaded with the right ask price. A failure means the data mapping, the regression data zip, or the symbol-mapping resolver returned a different ask price than the recorded golden value.

Source

Thrown at Algorithm.Python/OptionRenameRegressionAlgorithm.py:56

        if not self.portfolio.invested: 
            for kvp in slice.option_chains:
                chain = kvp.value
                if self.time.day == 28 and self.time.hour > 9 and self.time.minute > 0:
    
                    contracts = [i for i in sorted(chain, key=lambda x:x.expiry) 
                                         if i.right ==  OptionRight.CALL and 
                                            i.strike == 33 and
                                            i.expiry.date() == datetime(2013,8,17).date()]
                    if contracts:
                        # Buying option
                        contract = contracts[0]
                        self.buy(contract.symbol, 1)
                        # Buy the undelying stock
                        underlying_symbol = contract.symbol.underlying
                        self.buy (underlying_symbol, 100)
                        # check
                        if float(contract.ask_price) != 1.1:
                            raise ValueError("Regression test failed: current ask price was not loaded from NWSA backtest file and is not $1.1")
        elif self.time.day == 2 and self.time.hour > 14 and self.time.minute > 0:
            for kvp in slice.option_chains:
                chain = kvp.value
                self.liquidate()
                contracts = [i for i in sorted(chain, key=lambda x:x.expiry) 
                                        if i.right ==  OptionRight.CALL and 
                                           i.strike == 33 and
                                           i.expiry.date() == datetime(2013,8,17).date()]
            if contracts:
                contract = contracts[0]
                self.log("Bid Price" + str(contract.bid_price))
                if float(contract.bid_price) != 0.05:
                    raise ValueError("Regression test failed: current bid price was not loaded from FOXA file and is not $0.05")

    def on_order_event(self, order_event):
        self.log(str(order_event))

View on GitHub (pinned to d2c3659f87)

Solutions

Confirm the regression data zip for this algorithm is unchanged; if data was regenerated, update the golden expected value (1.1) to the new correct ask price.
Inspect the map/factor files under Lean/Data for TFCFA/NWSA to ensure the rename mapping resolves to the original NWSA backtest file.
Replace the brittle exact float equality (== 1.1) with a tolerance check (abs(ask - 1.1) < 1e-6) if the value is correct but floating-point representation changed.
Verify the SecurityIdentifier mapping resolver returns the same SID for the contract so the same quote row is loaded.

Example fix

// before
if float(contract.ask_price) != 1.1:
    raise ValueError('...not $1.1')
// after
if abs(float(contract.ask_price) - 1.1) > 1e-6:
    raise ValueError(f'ask price {contract.ask_price} != 1.1 after rename mapping')

Defensive patterns

Strategy: validation

Validate before calling

# Use a tolerance instead of exact float equality for loaded quote prices
ASK_GOLDEN = 1.1
if abs(float(contract.ask_price) - ASK_GOLDEN) > 1e-6:
    raise ValueError(f'NWSA ask {contract.ask_price} != {ASK_GOLDEN}')

Prevention

Never compare floats with == for prices loaded from data; use a tolerance.
Pin regression data zips so symbol-rename golden values stay stable.
When regenerating data, update golden constants in lockstep.
Validate map/factor files resolve the renamed ticker to the expected historical file.

When it happens

Trigger: Running the rename regression on 2013-06-28 after 09:00 when the option chain for TFCFA is available, selecting the CALL strike 33 / 2013-08-17 contract, and reading contract.ask_price when it is not float-equal to 1.1. Happens when the regression data package was regenerated, the symbol mapping changed, or a factor/rename file was edited.

Common situations: Updating the Lean regression data zips; changing the Symbol / SecurityIdentifier mapping logic; adding or editing a map file for NWSA/TFCFA/FOXA; floating the ask price through a normalization step that introduced rounding so 1.1 became 1.1000001.

Related errors

AI-assisted analysis of QuantConnect/Lean@d2c3659f87 (2026-08-13). Data as JSON: /api/errors/81c3598372f4ac8a. Report an issue: GitHub.