Unrealistic (exaggerated) slippage on ES in simulator?

[Johnny]

Hello all,

I've been trading ES with real money for a long time now and slippage is virtually never a problem, but now that I'm enrolled in a get-funded-program with one such company and trading in simulator initially I find that the slippage is extreme.

Yesterday, I experienced 2,5 point / 10 tick slippage on a stop order. I had a similar incident earlier in the week. With some other instrument this would probably be realistic, but not on ES trading 2-3 contracts.

My question is how does the simulator in Ninja account for slippage? Is it equal on all instruments? Is there always slippage or is there some kind of logic to determine when there's slippage and when there's not?

I'm not complaining - just trying to gain a better understanding. As for the get-funded program I'll just need to keep this in mind and use wider stops or limits to get out sooner.

Best regards,

Johnny

[NinjaTrader_ShawnB]

Hello Johnny,

Thank you for your post.

If you are using live data, submitting real orders to your evaluation account, your orders are filled based on the market price at the time of order submission. Slippage is not calculated or simulated in NinjaTrader unless you are backtesting an automated strategy.

Market orders will buy or sell immediately at the best available current price and are prone to slippage.

You can avoid slippage by using limit orders as these order types will fill at a guaranteed price. These orders are useful on low volatility instruments because they ensure you get filled at the price you specified or better. Take note that limit orders are not guaranteed to execute and may cause only partial fills.

Please do not hesitate to contact us for any other NinjaTrader inquiries you may have.

[Johnny]

Hi, Shawn,

I'm not sure if this answered my question. We can leave limit orders out of the question as I specifically refer to market orders as those are the only orders which will have slippage.

Having traded ES live for many years - slippage is virtually unheard of using market orders trading < 5 contracts in the US RTH session. 1 tick slippage can happen, but usually stop price = fill price. I remember once having a 6 tick slippage, but that's like 1 trade among thousands.

Friday, in simulator mode, I got 2,5 points / 10 ticks slippage on a trailing stop exiting a trade . See attached picture.

I was long from 4212,50 and placed a protective sell stop 2 ticks higher @ 4213,00 when price had moved in my favour. When triggered, this sell stop was executed 2,5 points lower giving me a 7,5 point loss.

[hedgeplay]

Yesterday, I experienced 2,5 point / 10 tick slippage on a stop order. I had a similar incident earlier in the week. With some other instrument this would probably be realistic, but not on ES trading 2-3 contracts.

Hey Johnny,

Just an FYI...

Whenever I see unexpected slippage or a rash of rejected orders in the Sim my default first suspect is market data lag (So in reality I am submitting orders far from the real market price).


In order to address this issue I use these general methods in all production work as well as sim


1) Add as BarsArray[1] a one tick data series to order against. Then just prior to every Order or ChangeOrder execute GetCurrentBid() or GetCurrentAsk() to get a more accurate current price.

https://ninjatrader.com/support/help...currentbid.htm
https://ninjatrader.com/support/help...currentask.htm

2) A few lines of code from an early strategy where I address data lag specifically.

.

OnBarUpdate()
{[INDENT]...[/INDENT][INDENT]// Frequently sync your PC clock to time.nist.gov to keep the line below accurate
TimeSpan dataLagTimeDiffTimeSpan = System.DateTime.Now - Times[1][0];

double dataLagTimeDiffTimeSpanDbl = dataLagTimeDiffTimeSpan.TotalSeconds;

// Define maxNTDataLagAllowedForEntryOrChanges in "Properties"
bool bDataLagIsOK = dataLagTimeDiffTimeSpanDbl < maxNTDataLagAllowedForEntryOrChanges;

// If Data Lag is Not within limit Entry is blocked.
if(bDataLagIsOK)
{[/INDENT][INDENT=2]EnterLong(.....)[/INDENT][INDENT]}
...[/INDENT]
 
   }


.

HedgePlay

[Johnny]

Thank you, Hedgeplay

I looked further into that slippage I experienced and see that it was indeed a very fast spike down, so maybe it was realistic after all. Just way more than I experience 90 % of the time with real money trading - hence why I was curious if it actually was realistic or exaggerated.

Johnny

[alrafer]

Hi Johnny , I just had a case of 4613.25 - 4609.75 = 3.5 points of slippage and delay in ES 03-22 (SP500), Minis, simulation account though. Needless to say that I lost the trade due to that - my market short order started too late and low.

Pretty puzzled now... Does it seem realistic - in other words, will I experience situations like this when I go real trading with thousands of dollars in ES (SP500 Minis)? Pretty scary...

Thanks for your opinion on this (also CCing hedgeplay ).

[alrafer]

In fairness, I see massive delay in this case as the issue: when I rerun this trade in playback it gets the order way earlier. When the trade happened in real time there was more than 2 bars (1000 ticks) of delay before my short order was filled. How could this be explained, NinjaTrader_ShawnB ?

Thanks a lot.

[NinjaTrader_PatrickG]

alrafer, what you're describing sounds like what we would call 'Chart rendering delay'. This can occur when the amount of incoming data paired with your workspace complexity/PC capabilities causes NinjaTrader charts to fall behind real-time data. With that said, I would not expect this to affect automated strategy executions, but only manually placed orders.

Please contact us through the platform by going to the Control Center > Help > Email Support window, describing what you're experiencing, then clicking Send so we may specifically troubleshoot what you're reporting using your NinjaTrader diagnostic files.