tracking_data_sync.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Here is a code demostrates how to merge tracking with event data with metrica sample data https://github.com/metrica-sports/sample-data as reference"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 136,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-08-07T07:16:21.618607Z",
     "start_time": "2022-08-07T07:16:21.606607Z"
    }
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import csv\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 257,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-08-07T07:35:48.243495Z",
     "start_time": "2022-08-07T07:35:48.228496Z"
    }
   },
   "outputs": [],
   "source": [
    "def tracking_data_process(DATADIR):\n",
    "\n",
    "    # First:  deal with file headers so that we can get the player names correct\n",
    "    csvfile =  open(DATADIR, 'r') # create a csv file reader\n",
    "    reader = csv.reader(csvfile) \n",
    "    teamnamefull = next(reader)[3].lower()\n",
    "    print(\"Reading team: %s\" % teamnamefull)\n",
    "    # construct column names\n",
    "    jerseys = [x for x in next(reader) if x != ''] # extract player jersey numbers from second row\n",
    "    columns = next(reader)\n",
    "    for i, j in enumerate(jerseys): # create x & y position column headers for each player\n",
    "        columns[i*2+3] = \"{}_x\".format(j)\n",
    "        columns[i*2+4] = \"{}_y\".format(j)\n",
    "    columns[-2] = \"ball_x\" # column headers for the x & y positions of the ball\n",
    "    columns[-1] = \"ball_y\"\n",
    "    # Second: read in tracking data and place into pandas Dataframe\n",
    "    tracking = pd.read_csv(DATADIR, names=columns, index_col='Frame', skiprows=3)\n",
    "    return tracking"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 284,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-08-07T09:56:52.908553Z",
     "start_time": "2022-08-07T09:56:52.616917Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Reading team: home\n"
     ]
    }
   ],
   "source": [
    "event_data = pd.read_csv('data/Sample_Game_1/Sample_Game_1_RawEventsData.csv')\n",
    "tracking_data = tracking_data_process('data/Sample_Game_1/Sample_Game_1_RawTrackingData_Home_Team.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 285,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-08-07T09:56:53.201929Z",
     "start_time": "2022-08-07T09:56:53.184366Z"
    },
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "event_data = event_data[event_data['Period'] == 2].reset_index()\n",
    "tracking_data = tracking_data[tracking_data['Period'] == 2].reset_index()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "While for sample data here is synchronized, imagine they are not e.g. offset by unknown amount for frames and we have to find the number out. Here the unknown frame is set to 30000"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 261,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-08-07T07:35:50.827323Z",
     "start_time": "2022-08-07T07:35:50.812323Z"
    }
   },
   "outputs": [],
   "source": [
    "unknown_frame_offset = 30000\n",
    "\n",
    "tracking_data['Frame'] += unknown_frame_offset"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Also tracking data will not have perfect agreement with event data location in real life so some random noise is added here."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 262,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-08-07T07:35:51.262118Z",
     "start_time": "2022-08-07T07:35:51.243236Z"
    }
   },
   "outputs": [],
   "source": [
    "tracking_data['ball_x'] += 0.005*np.random.random()\n",
    "tracking_data['ball_y'] += 0.005*np.random.random()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Then it's the main body of script: for every passes in event data, get the (starting) event x-y location, then we try to merge and iterate with data so the distance between the ball and event location is minimized. When we select a random frame, the distance between event and ball location will be essentially randomized, but for the frame that are synchronized, the distance will become minized.\n",
    "\n",
    "While event data and tracking data can be unknown frames apart, typically tracking data provider will start the game not too long after the first frame, and therefore it can be used as an initial guess instead of guessing every possible time."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 264,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-08-07T07:35:55.479153Z",
     "start_time": "2022-08-07T07:35:55.476152Z"
    }
   },
   "outputs": [],
   "source": [
    "intial_guess = tracking_data['Frame'][0] - event_data['Start Frame'][0]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "During the iteration, the distance between event and ball location is stored. Typically the min and max search frame can be adjusted so to decrease amount of run time. Certain threshold can also be added to stop the script early when minimum is found (not implemented here)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 271,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-08-07T07:37:26.299397Z",
     "start_time": "2022-08-07T07:37:03.900371Z"
    },
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "ball_dist = []\n",
    "min_search_frame = 0\n",
    "max_search_frame = 1000\n",
    "for i in range(min_search_frame,max_search_frame):\n",
    "    temp_tracking_data = tracking_data.copy()\n",
    "    \n",
    "    temp_tracking_data['mergeFrame'] = temp_tracking_data['Frame'] - i - intial_guess\n",
    "    merged_data = pd.merge(event_data[event_data.Type == 'PASS'][['Start Frame','Start X','Start Y']],temp_tracking_data[['mergeFrame','ball_x','ball_y']],left_on='Start Frame',right_on='mergeFrame')\n",
    "    ball_avg_dist = np.sqrt((merged_data['ball_x'] - merged_data['Start X']) ** 2 + (merged_data['ball_y'] - merged_data['Start Y']) ** 2)\n",
    "    ball_dist.append(np.mean(ball_avg_dist))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "When we plot the distance we can find there is a sharp minimum at a certain frame"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 272,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-08-07T07:37:32.431695Z",
     "start_time": "2022-08-07T07:37:32.335552Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[<matplotlib.lines.Line2D at 0x1f915c54c10>]"
      ]
     },
     "execution_count": 272,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.title('Distance between ball and event location vs frame')\n",
    "plt.plot(np.arange(len(ball_dist)),ball_dist)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "After adding back initial guess and guess range we can get the synchronized frame we want"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 275,
   "metadata": {
    "ExecuteTime": {
     "end_time": "2022-08-07T07:37:40.614310Z",
     "start_time": "2022-08-07T07:37:40.604310Z"
    }
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "30000"
      ]
     },
     "execution_count": 275,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "intial_guess + min_search_frame +  np.argmin(ball_dist)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}