Slippage is where our backtester calculates the realistic impact of your orders on the execution price you receive.

To me and others that impact seems very unrealistic in Q2.

So I decided to look at book data, 1 m bar data and see what Q2 default slippage model calculate.

Book data from BATS Exchange http://batstrading.com/bzx/book/TLT/
4/21/2016
Last updated 11:27:51

Orders Accepted
198,924
Total Volume
472,742

    SHARES  PRICE  
TOP OF BOOK  
Asks  
1,600   128.68  
2,800   128.67  
4,400   128.66  
1,600   128.65  
600 128.64  
Bids  
1,100   128.63  
1,600   128.62  
7,000   128.61  
3,600   128.6  
3,900   128.59

From the book I see that if I put market order for 11000 shares TLT
price per share will be 128.66 and order will be executed instantly.

Bar data from freestockcharts http://www.freestockcharts.com

    1m  TLT bar data  
11:28:00  
Open    128.64  
High    128.66  
Low 128.63  
Close   128.64  
Volume  14,500

Using Q2 default slippage model I calculate

Volume limit 363 at price 128.77

Possible volume in book is 30 times more then Q2 volume limit.
Q2 Price is greater than highest in the book for 0.09

Is that realistic?

To my mind by their nature market orders should not have any volume impact but only price impact and limit orders should have only volume impact and no price impact.

I do not want to use custom slippage model because it changing objective trading environment. And that environment do not depend on average algo writer.
So custom slippage model is some kind of self deception.

The right default slippage model should solve the problem.