Future contracts: Use the continuous aggregated contract or a specific contract for price analysis?

Hi Haakon

Since you are day trading, you will likely want the unadjusted raw price data from the ES contract with the largest open interest.

Running the cells below in a Research Notebook will load and display the one minute resolution ES data for 12/15/2022.  That date happens to be the day prior to the Expiration date for the December contract.  Because dataMappingMode = DataMappingMode.OpenInterest and dataNormalizationMode = DataNormalizationMode.Raw, you will get the actual unadjusted price data for the March 23 contract because it has a higher open interest than the soon to be expired Dec 22 contract.

Note also that my “get_ES_Hist” function assumes you want the extended trading hours data as well, so the data will start at 6pm ET the prior evening and run through the end date.

Hope this helps,

MikeJ

qb = QuantBook()

def get_ES_Hist(sdt, edt):   
    start_dt = datetime.combine(sdt-timedelta(1), time(18, 0))
    end_dt   = datetime.combine(edt, time(18, 0))


    print(f'History start: {start_dt.strftime("%A, %d %B %Y %H:%M:%S")}')
    print(f'      through: {  end_dt.strftime("%A, %d %B %Y %H:%M:%S")}')

    # Rolls over when the following back month contract has 
    # a higher open interest that the current front month
    
    symbl_es = qb.AddFuture(Futures.Indices.SP500EMini).Symbol
        
    h = qb.History(
        TradeBar,
        tickers=[symbl_es], 
        start=start_dt, 
        end=end_dt, 
        resolution=Resolution.Minute, 
        fillForward=False, 
        extendedMarketHours=True, 
        dataMappingMode=DataMappingMode.OpenInterest, 
        dataNormalizationMode=DataNormalizationMode.Raw, 
        contractDepthOffset=0)
    return h
    
start_date = datetime(2022,12,15)
end_date = datetime(2022,12,15)

history = get_ES_Hist(start_date, end_date)
history.index = history.index.droplevel([0,1])   

import plotly.graph_objects as go 
candlestick = go.Candlestick(x=history.index,
                             open=history['open'],
                             high=history['high'],
                             low=history['low'],
                             close=history['close'],
                             name='Candlestick Bar')

layout = go.Layout(title=go.layout.Title(text=f'Continuous ES'),
                            xaxis_title='Date',
                            yaxis_title='Price',
                            xaxis_rangeslider_visible=False)
                            
fig = go.Figure(data=[candlestick], layout=layout)
fig.show()

MikeJ  Thanks a lot for the reply! 😊 I have some follow-up questions:

1. Is it recommended to go for the contract with the highest open interest when day trading as it tends to be more volatile?
2. You title your plot “Continuous ES” but in this case is it right that you have selected the contract with the highest open interest which is different from the continuous symbol you get from subscribing to the E-mini in an algorithm via: self.AddFuture(Futures.Indices.SP500EMini) ? 
3. I extended the code to plot an additional chart using a different data mapping mode. The charts move identically, but on different price levels. Is that the case for all contracts in a future instrument, and why is that?

Code:

qb = QuantBook()

def get_ES_Hist(sdt, edt):   
    start_dt = datetime.combine(sdt-timedelta(1), time(18, 0))
    end_dt   = datetime.combine(edt, time(18, 0))


    print(f'History start: {start_dt.strftime("%A, %d %B %Y %H:%M:%S")}')
    print(f'      through: {  end_dt.strftime("%A, %d %B %Y %H:%M:%S")}')

    # Rolls over when the following back month contract has 
    # a higher open interest that the current front month
    
    symbl_es = qb.AddFuture(Futures.Indices.SP500EMini).Symbol
        
    h1 = qb.History(
        TradeBar,
        tickers=[symbl_es], 
        start=start_dt, 
        end=end_dt, 
        resolution=Resolution.Minute, 
        fillForward=False, 
        extendedMarketHours=True, 
        dataMappingMode=DataMappingMode.OpenInterest, 
        dataNormalizationMode=DataNormalizationMode.Raw, 
        contractDepthOffset=0)
    
    h2 = qb.History(
        TradeBar,
        tickers=[symbl_es], 
        start=start_dt, 
        end=end_dt, 
        resolution=Resolution.Minute, 
        fillForward=False, 
        extendedMarketHours=True, 
        dataMappingMode=DataMappingMode.LastTradingDay, 
        dataNormalizationMode=DataNormalizationMode.Raw, 
        contractDepthOffset=0)
    return h1, h2
    
start_date = datetime(2022,12,15)
end_date = datetime(2022,12,15)

history, history2 = get_ES_Hist(start_date, end_date)
history.index = history.index.droplevel([0,1])  
history2.index = history2.index.droplevel([0,1])  

import plotly.graph_objects as go 
candlestick = go.Candlestick(x=history.index,
                             open=history['open'],
                             high=history['high'],
                             low=history['low'],
                             close=history['close'],
                             name='Candlestick Bar 1')

candlestick2 = go.Candlestick(x=history.index,
                             open=history2['open'],
                             high=history2['high'],
                             low=history2['low'],
                             close=history2['close'],
                             name='Candlestick Bar 2')

layout = go.Layout(title=go.layout.Title(text=f'Continuous ES'),
                            xaxis_title='Date',
                            yaxis_title='Price',
                            xaxis_rangeslider_visible=False)
                            
fig = go.Figure(data=[candlestick, candlestick2], layout=layout)
fig.show()

1. The contract with the highest open interest is most likely to be the one with the most liquidity, so typically you get a smaller bid/ask spread (better fills).
2. Correct, the “raw” (unadjusted) and highest open interest on the 15th is the Mar 23 contract.  For back-testing strategies that only depend on data gathered on the trade day, this is all you need.  If back-testing a strategy that depends on older data, you will probably want to select a different mode besides “raw” so the flow is continuous when the strategy rolls to the newer contract.  I think the default is roll over on expiration day, but I'm not sure…
3. The short answer is, the contract price at any given moment is supposed to reflect the “expected” price at expiration.  There are other factors that influence the price, but for example, since on 12/15/22 the December contract expires the next day you would expect it's price to be close to the index (should match it when it expires).  But the March contract price reflects what the market thinks the index will be in March of 2023.  This is terribly simplified and there's lot's of good information about how future contracts work online.

Best,

MikeJ

B9 Mike Thanks again!

1. What about using DataMappingMode.OpenInterestAnual (Description: The contract maps when any of the back month contracts of the next year have a higher volume that the current front month (3)) ? Is that even better than the regular Open interest in terms of liquidity?
2. Am I correct in that I only need to subscribe to the price data of the continuous future, and simply need to reference the contract when entering a trade? I don't need to build my indicators around the price of a specific contract? Because the money (margin) required to buy 1 future contract will be the same regardless of which of the specific contracts I choose to buy? And the price of a tick and point is the same regardless of the specific contract as well? The only thing that changes is the amount of leverage used?