diff --git "a/nbs/confidence_analysis.ipynb" "b/nbs/confidence_analysis.ipynb"
deleted file mode 100644--- "a/nbs/confidence_analysis.ipynb"
+++ /dev/null
@@ -1,1032 +0,0 @@
-{
- "cells": [
-  {
-   "cell_type": "code",
-   "execution_count": 3,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "import pandas as pd\n",
-    "import matplotlib.pyplot as plt\n",
-    "import seaborn as sns\n",
-    "sns.set_style(\"darkgrid\")"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 4,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "import warnings\n",
-    "warnings.filterwarnings('ignore')"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 5,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "tools = pd.read_parquet('../data/tools.parquet')"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 6,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "inc_tools = [\n",
-    "    'prediction-online', \n",
-    "    'prediction-offline', \n",
-    "    'claude-prediction-online', \n",
-    "    'claude-prediction-offline', \n",
-    "    'prediction-offline-sme',\n",
-    "    'prediction-online-sme',\n",
-    "    'prediction-request-rag',\n",
-    "    'prediction-request-reasoning',\n",
-    "    'prediction-url-cot-claude', \n",
-    "    'prediction-request-rag-claude',\n",
-    "    'prediction-request-reasoning-claude'\n",
-    "]\n",
-    "# include only tools that are in inc_tools\n",
-    "tools_inc = tools[tools['tool'].isin(inc_tools)]"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 7,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "# only include non error requests\n",
-    "tools_non_error = tools_inc[tools_inc['error'] != 1]\n",
-    "tools_non_error['currentAnswer'].replace('no', 'No', inplace=True)\n",
-    "tools_non_error['currentAnswer'].replace('yes', 'Yes', inplace=True)\n",
-    "tools_non_error = tools_non_error[tools_non_error['currentAnswer'].isin(['Yes', 'No'])]\n",
-    "tools_non_error = tools_non_error[tools_non_error['vote'].isin(['Yes', 'No'])]"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 8,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "tools_non_error['win'] = tools_non_error['currentAnswer'] == tools_non_error['vote']\n",
-    "tools_non_error['win'] = tools_non_error['win'].astype(int)"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 8,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<div>\n",
-       "<style scoped>\n",
-       "    .dataframe tbody tr th:only-of-type {\n",
-       "        vertical-align: middle;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe tbody tr th {\n",
-       "        vertical-align: top;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "</style>\n",
-       "<table border=\"1\" class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr style=\"text-align: right;\">\n",
-       "      <th></th>\n",
-       "      <th>request_id</th>\n",
-       "      <th>request_block</th>\n",
-       "      <th>prompt_request</th>\n",
-       "      <th>tool</th>\n",
-       "      <th>nonce</th>\n",
-       "      <th>trader_address</th>\n",
-       "      <th>deliver_block</th>\n",
-       "      <th>error</th>\n",
-       "      <th>error_message</th>\n",
-       "      <th>prompt_response</th>\n",
-       "      <th>...</th>\n",
-       "      <th>confidence</th>\n",
-       "      <th>info_utility</th>\n",
-       "      <th>vote</th>\n",
-       "      <th>win_probability</th>\n",
-       "      <th>title</th>\n",
-       "      <th>currentAnswer</th>\n",
-       "      <th>request_time</th>\n",
-       "      <th>request_month_year</th>\n",
-       "      <th>request_month_year_week</th>\n",
-       "      <th>win</th>\n",
-       "    </tr>\n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    <tr>\n",
-       "      <th>0</th>\n",
-       "      <td>3006386814233319828691780626079673319114537324...</td>\n",
-       "      <td>29624577</td>\n",
-       "      <td>With the given question \"Will the Bored Ape Ya...</td>\n",
-       "      <td>prediction-online</td>\n",
-       "      <td>f53a0449-fd88-4b19-af92-419818f83b74</td>\n",
-       "      <td>0x44d97681a1d4d86D8Ddb7D960b063E22FD9DB490</td>\n",
-       "      <td>29624613</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>None</td>\n",
-       "      <td>None</td>\n",
-       "      <td>...</td>\n",
-       "      <td>0.8</td>\n",
-       "      <td>0.9</td>\n",
-       "      <td>Yes</td>\n",
-       "      <td>0.60</td>\n",
-       "      <td>Will the Bored Ape Yacht Club still be blockin...</td>\n",
-       "      <td>No</td>\n",
-       "      <td>2023-08-24 16:04:50</td>\n",
-       "      <td>2023-08</td>\n",
-       "      <td>2023-08-21/2023-08-27</td>\n",
-       "      <td>0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>1</th>\n",
-       "      <td>3331123099573261158036411767774642098138230831...</td>\n",
-       "      <td>29624616</td>\n",
-       "      <td>With the given question \"Will the Xbox 360 sto...</td>\n",
-       "      <td>prediction-online</td>\n",
-       "      <td>1a5a576c-2430-4e09-a1ec-a8166da356d4</td>\n",
-       "      <td>0x44d97681a1d4d86D8Ddb7D960b063E22FD9DB490</td>\n",
-       "      <td>29625988</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>None</td>\n",
-       "      <td>None</td>\n",
-       "      <td>...</td>\n",
-       "      <td>0.7</td>\n",
-       "      <td>0.6</td>\n",
-       "      <td>No</td>\n",
-       "      <td>0.80</td>\n",
-       "      <td>Will the Xbox 360 store close before 24 August...</td>\n",
-       "      <td>No</td>\n",
-       "      <td>2023-08-24 16:08:20</td>\n",
-       "      <td>2023-08</td>\n",
-       "      <td>2023-08-21/2023-08-27</td>\n",
-       "      <td>1</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>4</th>\n",
-       "      <td>8876870293926357350592315669265884828102242482...</td>\n",
-       "      <td>29625975</td>\n",
-       "      <td>With the given question \"Will Arctic sea ice r...</td>\n",
-       "      <td>prediction-online</td>\n",
-       "      <td>74ba8707-a7b7-4979-8845-de7d4400b011</td>\n",
-       "      <td>0x44d97681a1d4d86D8Ddb7D960b063E22FD9DB490</td>\n",
-       "      <td>29626031</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>None</td>\n",
-       "      <td>None</td>\n",
-       "      <td>...</td>\n",
-       "      <td>0.8</td>\n",
-       "      <td>0.7</td>\n",
-       "      <td>Yes</td>\n",
-       "      <td>0.60</td>\n",
-       "      <td>Will Arctic sea ice reach a new lowest level b...</td>\n",
-       "      <td>No</td>\n",
-       "      <td>2023-08-24 18:05:35</td>\n",
-       "      <td>2023-08</td>\n",
-       "      <td>2023-08-21/2023-08-27</td>\n",
-       "      <td>0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>5</th>\n",
-       "      <td>2909784377514364366708664729310496697567238257...</td>\n",
-       "      <td>29626002</td>\n",
-       "      <td>With the given question \"Will Amazon have a ne...</td>\n",
-       "      <td>prediction-online</td>\n",
-       "      <td>41e37a98-b091-46bb-a78a-10865ac20600</td>\n",
-       "      <td>0xDF5d21397543Eb0fB47aFc616073cD922E020635</td>\n",
-       "      <td>29626020</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>None</td>\n",
-       "      <td>None</td>\n",
-       "      <td>...</td>\n",
-       "      <td>0.8</td>\n",
-       "      <td>0.6</td>\n",
-       "      <td>Yes</td>\n",
-       "      <td>0.65</td>\n",
-       "      <td>Will Amazon have a new Senior Vice President o...</td>\n",
-       "      <td>No</td>\n",
-       "      <td>2023-08-24 18:08:00</td>\n",
-       "      <td>2023-08</td>\n",
-       "      <td>2023-08-21/2023-08-27</td>\n",
-       "      <td>0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>6</th>\n",
-       "      <td>1027551294836394984613821404170252168134524127...</td>\n",
-       "      <td>29626041</td>\n",
-       "      <td>With the given question \"Will the 'Jalapeño Ga...</td>\n",
-       "      <td>prediction-online</td>\n",
-       "      <td>c7e9c3c1-dc20-4126-b198-c10bf3257d1b</td>\n",
-       "      <td>0x44d97681a1d4d86D8Ddb7D960b063E22FD9DB490</td>\n",
-       "      <td>29626076</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>None</td>\n",
-       "      <td>None</td>\n",
-       "      <td>...</td>\n",
-       "      <td>0.8</td>\n",
-       "      <td>0.6</td>\n",
-       "      <td>Yes</td>\n",
-       "      <td>0.70</td>\n",
-       "      <td>Will the 'Jalapeño Gate' controversy be resolv...</td>\n",
-       "      <td>No</td>\n",
-       "      <td>2023-08-24 18:11:20</td>\n",
-       "      <td>2023-08</td>\n",
-       "      <td>2023-08-21/2023-08-27</td>\n",
-       "      <td>0</td>\n",
-       "    </tr>\n",
-       "  </tbody>\n",
-       "</table>\n",
-       "<p>5 rows × 23 columns</p>\n",
-       "</div>"
-      ],
-      "text/plain": [
-       "                                          request_id  request_block  \\\n",
-       "0  3006386814233319828691780626079673319114537324...       29624577   \n",
-       "1  3331123099573261158036411767774642098138230831...       29624616   \n",
-       "4  8876870293926357350592315669265884828102242482...       29625975   \n",
-       "5  2909784377514364366708664729310496697567238257...       29626002   \n",
-       "6  1027551294836394984613821404170252168134524127...       29626041   \n",
-       "\n",
-       "                                      prompt_request               tool  \\\n",
-       "0  With the given question \"Will the Bored Ape Ya...  prediction-online   \n",
-       "1  With the given question \"Will the Xbox 360 sto...  prediction-online   \n",
-       "4  With the given question \"Will Arctic sea ice r...  prediction-online   \n",
-       "5  With the given question \"Will Amazon have a ne...  prediction-online   \n",
-       "6  With the given question \"Will the 'Jalapeño Ga...  prediction-online   \n",
-       "\n",
-       "                                  nonce  \\\n",
-       "0  f53a0449-fd88-4b19-af92-419818f83b74   \n",
-       "1  1a5a576c-2430-4e09-a1ec-a8166da356d4   \n",
-       "4  74ba8707-a7b7-4979-8845-de7d4400b011   \n",
-       "5  41e37a98-b091-46bb-a78a-10865ac20600   \n",
-       "6  c7e9c3c1-dc20-4126-b198-c10bf3257d1b   \n",
-       "\n",
-       "                               trader_address  deliver_block  error  \\\n",
-       "0  0x44d97681a1d4d86D8Ddb7D960b063E22FD9DB490       29624613    0.0   \n",
-       "1  0x44d97681a1d4d86D8Ddb7D960b063E22FD9DB490       29625988    0.0   \n",
-       "4  0x44d97681a1d4d86D8Ddb7D960b063E22FD9DB490       29626031    0.0   \n",
-       "5  0xDF5d21397543Eb0fB47aFc616073cD922E020635       29626020    0.0   \n",
-       "6  0x44d97681a1d4d86D8Ddb7D960b063E22FD9DB490       29626076    0.0   \n",
-       "\n",
-       "  error_message prompt_response  ... confidence  info_utility  vote  \\\n",
-       "0          None            None  ...        0.8           0.9   Yes   \n",
-       "1          None            None  ...        0.7           0.6    No   \n",
-       "4          None            None  ...        0.8           0.7   Yes   \n",
-       "5          None            None  ...        0.8           0.6   Yes   \n",
-       "6          None            None  ...        0.8           0.6   Yes   \n",
-       "\n",
-       "   win_probability                                              title  \\\n",
-       "0             0.60  Will the Bored Ape Yacht Club still be blockin...   \n",
-       "1             0.80  Will the Xbox 360 store close before 24 August...   \n",
-       "4             0.60  Will Arctic sea ice reach a new lowest level b...   \n",
-       "5             0.65  Will Amazon have a new Senior Vice President o...   \n",
-       "6             0.70  Will the 'Jalapeño Gate' controversy be resolv...   \n",
-       "\n",
-       "  currentAnswer         request_time request_month_year  \\\n",
-       "0            No  2023-08-24 16:04:50            2023-08   \n",
-       "1            No  2023-08-24 16:08:20            2023-08   \n",
-       "4            No  2023-08-24 18:05:35            2023-08   \n",
-       "5            No  2023-08-24 18:08:00            2023-08   \n",
-       "6            No  2023-08-24 18:11:20            2023-08   \n",
-       "\n",
-       "  request_month_year_week win  \n",
-       "0   2023-08-21/2023-08-27   0  \n",
-       "1   2023-08-21/2023-08-27   1  \n",
-       "4   2023-08-21/2023-08-27   0  \n",
-       "5   2023-08-21/2023-08-27   0  \n",
-       "6   2023-08-21/2023-08-27   0  \n",
-       "\n",
-       "[5 rows x 23 columns]"
-      ]
-     },
-     "execution_count": 8,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "tools_non_error.head()"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 19,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<div>\n",
-       "<style scoped>\n",
-       "    .dataframe tbody tr th:only-of-type {\n",
-       "        vertical-align: middle;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe tbody tr th {\n",
-       "        vertical-align: top;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "</style>\n",
-       "<table border=\"1\" class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr style=\"text-align: right;\">\n",
-       "      <th></th>\n",
-       "      <th>p_yes</th>\n",
-       "      <th>p_no</th>\n",
-       "      <th>win_probability</th>\n",
-       "    </tr>\n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    <tr>\n",
-       "      <th>0</th>\n",
-       "      <td>0.60</td>\n",
-       "      <td>0.40</td>\n",
-       "      <td>0.60</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>1</th>\n",
-       "      <td>0.20</td>\n",
-       "      <td>0.80</td>\n",
-       "      <td>0.80</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>4</th>\n",
-       "      <td>0.60</td>\n",
-       "      <td>0.40</td>\n",
-       "      <td>0.60</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>5</th>\n",
-       "      <td>0.65</td>\n",
-       "      <td>0.35</td>\n",
-       "      <td>0.65</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>6</th>\n",
-       "      <td>0.70</td>\n",
-       "      <td>0.30</td>\n",
-       "      <td>0.70</td>\n",
-       "    </tr>\n",
-       "  </tbody>\n",
-       "</table>\n",
-       "</div>"
-      ],
-      "text/plain": [
-       "   p_yes  p_no  win_probability\n",
-       "0   0.60  0.40             0.60\n",
-       "1   0.20  0.80             0.80\n",
-       "4   0.60  0.40             0.60\n",
-       "5   0.65  0.35             0.65\n",
-       "6   0.70  0.30             0.70"
-      ]
-     },
-     "execution_count": 19,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "tools_non_error[[\"p_yes\", \"p_no\", \"win_probability\"]].head()"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "# Confidence analysis per Trader answer type"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 9,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "Text(0.5, 1.0, 'Distribution of confidence on the winning trades')"
-      ]
-     },
-     "execution_count": 9,
-     "metadata": {},
-     "output_type": "execute_result"
-    },
-    {
-     "data": {
-      "image/png": "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",
-      "text/plain": [
-       "<Figure size 500x500 with 1 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
-   "source": [
-    "sns.displot(data=tools_non_error[tools_non_error[\"win\"]==1], x=\"confidence\", kde=True)\n",
-    "plt.xlabel('Confidence percentage')\n",
-    "plt.ylabel('Count')\n",
-    "plt.title('Distribution of confidence on the winning trades')"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 9,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "winning_trades = tools_non_error[tools_non_error[\"win\"]==1]\n",
-    "non_winning_trades = tools_non_error[tools_non_error[\"win\"]==0]"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 14,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "Text(0.5, 1.0, 'Distribution of confidence on the NON-winning trades')"
-      ]
-     },
-     "execution_count": 14,
-     "metadata": {},
-     "output_type": "execute_result"
-    },
-    {
-     "data": {
-      "image/png": "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",
-      "text/plain": [
-       "<Figure size 500x500 with 1 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
-   "source": [
-    "sns.displot(data=tools_non_error[tools_non_error[\"win\"]==0], x=\"confidence\", kde=True)\n",
-    "plt.xlabel('Confidence percentage')\n",
-    "plt.ylabel('Count')\n",
-    "plt.title('Distribution of confidence on the NON-winning trades')"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 15,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "win\n",
-       "1    109540\n",
-       "0     91498\n",
-       "Name: count, dtype: int64"
-      ]
-     },
-     "execution_count": 15,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "tools_non_error.win.value_counts()"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 15,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<div>\n",
-       "<style scoped>\n",
-       "    .dataframe tbody tr th:only-of-type {\n",
-       "        vertical-align: middle;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe tbody tr th {\n",
-       "        vertical-align: top;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "</style>\n",
-       "<table border=\"1\" class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr style=\"text-align: right;\">\n",
-       "      <th></th>\n",
-       "      <th>request_block</th>\n",
-       "      <th>deliver_block</th>\n",
-       "      <th>error</th>\n",
-       "      <th>p_yes</th>\n",
-       "      <th>p_no</th>\n",
-       "      <th>confidence</th>\n",
-       "      <th>info_utility</th>\n",
-       "      <th>win_probability</th>\n",
-       "      <th>win</th>\n",
-       "    </tr>\n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    <tr>\n",
-       "      <th>count</th>\n",
-       "      <td>1.095400e+05</td>\n",
-       "      <td>1.095400e+05</td>\n",
-       "      <td>109540.0</td>\n",
-       "      <td>109540.000000</td>\n",
-       "      <td>109540.000000</td>\n",
-       "      <td>109540.000000</td>\n",
-       "      <td>109540.000000</td>\n",
-       "      <td>109540.000000</td>\n",
-       "      <td>109540.0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>mean</th>\n",
-       "      <td>3.252190e+07</td>\n",
-       "      <td>3.252205e+07</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>0.453541</td>\n",
-       "      <td>0.546459</td>\n",
-       "      <td>0.749623</td>\n",
-       "      <td>0.643063</td>\n",
-       "      <td>0.713424</td>\n",
-       "      <td>1.0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>std</th>\n",
-       "      <td>1.246011e+06</td>\n",
-       "      <td>1.245733e+06</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>0.236433</td>\n",
-       "      <td>0.236433</td>\n",
-       "      <td>0.116751</td>\n",
-       "      <td>0.234643</td>\n",
-       "      <td>0.111842</td>\n",
-       "      <td>0.0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>min</th>\n",
-       "      <td>2.902952e+07</td>\n",
-       "      <td>2.902953e+07</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>0.000000</td>\n",
-       "      <td>0.000000</td>\n",
-       "      <td>0.000000</td>\n",
-       "      <td>0.000000</td>\n",
-       "      <td>0.502000</td>\n",
-       "      <td>1.0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>25%</th>\n",
-       "      <td>3.157338e+07</td>\n",
-       "      <td>3.157339e+07</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>0.300000</td>\n",
-       "      <td>0.350000</td>\n",
-       "      <td>0.700000</td>\n",
-       "      <td>0.500000</td>\n",
-       "      <td>0.600000</td>\n",
-       "      <td>1.0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>50%</th>\n",
-       "      <td>3.300844e+07</td>\n",
-       "      <td>3.300848e+07</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>0.400000</td>\n",
-       "      <td>0.600000</td>\n",
-       "      <td>0.800000</td>\n",
-       "      <td>0.700000</td>\n",
-       "      <td>0.700000</td>\n",
-       "      <td>1.0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>75%</th>\n",
-       "      <td>3.348162e+07</td>\n",
-       "      <td>3.348164e+07</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>0.650000</td>\n",
-       "      <td>0.700000</td>\n",
-       "      <td>0.800000</td>\n",
-       "      <td>0.800000</td>\n",
-       "      <td>0.800000</td>\n",
-       "      <td>1.0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>max</th>\n",
-       "      <td>3.406412e+07</td>\n",
-       "      <td>3.406415e+07</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>1.000000</td>\n",
-       "      <td>1.000000</td>\n",
-       "      <td>1.000000</td>\n",
-       "      <td>1.000000</td>\n",
-       "      <td>1.000000</td>\n",
-       "      <td>1.0</td>\n",
-       "    </tr>\n",
-       "  </tbody>\n",
-       "</table>\n",
-       "</div>"
-      ],
-      "text/plain": [
-       "       request_block  deliver_block     error          p_yes           p_no  \\\n",
-       "count   1.095400e+05   1.095400e+05  109540.0  109540.000000  109540.000000   \n",
-       "mean    3.252190e+07   3.252205e+07       0.0       0.453541       0.546459   \n",
-       "std     1.246011e+06   1.245733e+06       0.0       0.236433       0.236433   \n",
-       "min     2.902952e+07   2.902953e+07       0.0       0.000000       0.000000   \n",
-       "25%     3.157338e+07   3.157339e+07       0.0       0.300000       0.350000   \n",
-       "50%     3.300844e+07   3.300848e+07       0.0       0.400000       0.600000   \n",
-       "75%     3.348162e+07   3.348164e+07       0.0       0.650000       0.700000   \n",
-       "max     3.406412e+07   3.406415e+07       0.0       1.000000       1.000000   \n",
-       "\n",
-       "          confidence   info_utility  win_probability       win  \n",
-       "count  109540.000000  109540.000000    109540.000000  109540.0  \n",
-       "mean        0.749623       0.643063         0.713424       1.0  \n",
-       "std         0.116751       0.234643         0.111842       0.0  \n",
-       "min         0.000000       0.000000         0.502000       1.0  \n",
-       "25%         0.700000       0.500000         0.600000       1.0  \n",
-       "50%         0.800000       0.700000         0.700000       1.0  \n",
-       "75%         0.800000       0.800000         0.800000       1.0  \n",
-       "max         1.000000       1.000000         1.000000       1.0  "
-      ]
-     },
-     "execution_count": 15,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "winning_trades.describe()"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 17,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "count    109540.000000\n",
-       "mean          0.749623\n",
-       "std           0.116751\n",
-       "min           0.000000\n",
-       "25%           0.700000\n",
-       "50%           0.800000\n",
-       "75%           0.800000\n",
-       "max           1.000000\n",
-       "Name: confidence, dtype: float64"
-      ]
-     },
-     "execution_count": 17,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "winning_trades.confidence.describe()"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 16,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/html": [
-       "<div>\n",
-       "<style scoped>\n",
-       "    .dataframe tbody tr th:only-of-type {\n",
-       "        vertical-align: middle;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe tbody tr th {\n",
-       "        vertical-align: top;\n",
-       "    }\n",
-       "\n",
-       "    .dataframe thead th {\n",
-       "        text-align: right;\n",
-       "    }\n",
-       "</style>\n",
-       "<table border=\"1\" class=\"dataframe\">\n",
-       "  <thead>\n",
-       "    <tr style=\"text-align: right;\">\n",
-       "      <th></th>\n",
-       "      <th>request_block</th>\n",
-       "      <th>deliver_block</th>\n",
-       "      <th>error</th>\n",
-       "      <th>p_yes</th>\n",
-       "      <th>p_no</th>\n",
-       "      <th>confidence</th>\n",
-       "      <th>info_utility</th>\n",
-       "      <th>win_probability</th>\n",
-       "      <th>win</th>\n",
-       "    </tr>\n",
-       "  </thead>\n",
-       "  <tbody>\n",
-       "    <tr>\n",
-       "      <th>count</th>\n",
-       "      <td>9.149800e+04</td>\n",
-       "      <td>9.149800e+04</td>\n",
-       "      <td>91498.0</td>\n",
-       "      <td>91498.000000</td>\n",
-       "      <td>91498.000000</td>\n",
-       "      <td>91498.000000</td>\n",
-       "      <td>91498.000000</td>\n",
-       "      <td>91498.000000</td>\n",
-       "      <td>91498.0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>mean</th>\n",
-       "      <td>3.233187e+07</td>\n",
-       "      <td>3.233206e+07</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>0.523302</td>\n",
-       "      <td>0.476698</td>\n",
-       "      <td>0.761938</td>\n",
-       "      <td>0.644957</td>\n",
-       "      <td>0.695807</td>\n",
-       "      <td>0.0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>std</th>\n",
-       "      <td>1.263300e+06</td>\n",
-       "      <td>1.262970e+06</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>0.219429</td>\n",
-       "      <td>0.219429</td>\n",
-       "      <td>0.107188</td>\n",
-       "      <td>0.218935</td>\n",
-       "      <td>0.101741</td>\n",
-       "      <td>0.0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>min</th>\n",
-       "      <td>2.902954e+07</td>\n",
-       "      <td>2.902955e+07</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>0.000000</td>\n",
-       "      <td>0.000000</td>\n",
-       "      <td>0.000000</td>\n",
-       "      <td>0.000000</td>\n",
-       "      <td>0.502000</td>\n",
-       "      <td>0.0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>25%</th>\n",
-       "      <td>3.140342e+07</td>\n",
-       "      <td>3.140344e+07</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>0.300000</td>\n",
-       "      <td>0.300000</td>\n",
-       "      <td>0.700000</td>\n",
-       "      <td>0.500000</td>\n",
-       "      <td>0.600000</td>\n",
-       "      <td>0.0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>50%</th>\n",
-       "      <td>3.277588e+07</td>\n",
-       "      <td>3.277591e+07</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>0.600000</td>\n",
-       "      <td>0.400000</td>\n",
-       "      <td>0.800000</td>\n",
-       "      <td>0.700000</td>\n",
-       "      <td>0.700000</td>\n",
-       "      <td>0.0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>75%</th>\n",
-       "      <td>3.330261e+07</td>\n",
-       "      <td>3.330264e+07</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>0.700000</td>\n",
-       "      <td>0.700000</td>\n",
-       "      <td>0.800000</td>\n",
-       "      <td>0.800000</td>\n",
-       "      <td>0.800000</td>\n",
-       "      <td>0.0</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <th>max</th>\n",
-       "      <td>3.406362e+07</td>\n",
-       "      <td>3.406364e+07</td>\n",
-       "      <td>0.0</td>\n",
-       "      <td>1.000000</td>\n",
-       "      <td>1.000000</td>\n",
-       "      <td>1.000000</td>\n",
-       "      <td>1.000000</td>\n",
-       "      <td>1.000000</td>\n",
-       "      <td>0.0</td>\n",
-       "    </tr>\n",
-       "  </tbody>\n",
-       "</table>\n",
-       "</div>"
-      ],
-      "text/plain": [
-       "       request_block  deliver_block    error         p_yes          p_no  \\\n",
-       "count   9.149800e+04   9.149800e+04  91498.0  91498.000000  91498.000000   \n",
-       "mean    3.233187e+07   3.233206e+07      0.0      0.523302      0.476698   \n",
-       "std     1.263300e+06   1.262970e+06      0.0      0.219429      0.219429   \n",
-       "min     2.902954e+07   2.902955e+07      0.0      0.000000      0.000000   \n",
-       "25%     3.140342e+07   3.140344e+07      0.0      0.300000      0.300000   \n",
-       "50%     3.277588e+07   3.277591e+07      0.0      0.600000      0.400000   \n",
-       "75%     3.330261e+07   3.330264e+07      0.0      0.700000      0.700000   \n",
-       "max     3.406362e+07   3.406364e+07      0.0      1.000000      1.000000   \n",
-       "\n",
-       "         confidence  info_utility  win_probability      win  \n",
-       "count  91498.000000  91498.000000     91498.000000  91498.0  \n",
-       "mean       0.761938      0.644957         0.695807      0.0  \n",
-       "std        0.107188      0.218935         0.101741      0.0  \n",
-       "min        0.000000      0.000000         0.502000      0.0  \n",
-       "25%        0.700000      0.500000         0.600000      0.0  \n",
-       "50%        0.800000      0.700000         0.700000      0.0  \n",
-       "75%        0.800000      0.800000         0.800000      0.0  \n",
-       "max        1.000000      1.000000         1.000000      0.0  "
-      ]
-     },
-     "execution_count": 16,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "non_winning_trades.describe()"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 18,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "text/plain": [
-       "count    91498.000000\n",
-       "mean         0.761938\n",
-       "std          0.107188\n",
-       "min          0.000000\n",
-       "25%          0.700000\n",
-       "50%          0.800000\n",
-       "75%          0.800000\n",
-       "max          1.000000\n",
-       "Name: confidence, dtype: float64"
-      ]
-     },
-     "execution_count": 18,
-     "metadata": {},
-     "output_type": "execute_result"
-    }
-   ],
-   "source": [
-    "non_winning_trades.confidence.describe()"
-   ]
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "This means on average the non winning trades show a higher confidence."
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "metadata": {},
-   "outputs": [],
-   "source": []
-  },
-  {
-   "cell_type": "markdown",
-   "metadata": {},
-   "source": [
-    "# Correlation between confidence and win_probability"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 21,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "image/png": "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",
-      "text/plain": [
-       "<Figure size 640x480 with 2 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
-   "source": [
-    "\n",
-    "corr_matrix = tools_non_error[[\"confidence\", \"win_probability\"]].corr().round(2)\n",
-    "sns.heatmap(corr_matrix, annot=True)\n",
-    "plt.show()"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 22,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "image/png": "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",
-      "text/plain": [
-       "<Figure size 640x480 with 2 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
-   "source": [
-    "corr_matrix = winning_trades[[\"confidence\", \"win_probability\"]].corr().round(2)\n",
-    "sns.heatmap(corr_matrix, annot=True)\n",
-    "plt.show()"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": 23,
-   "metadata": {},
-   "outputs": [
-    {
-     "data": {
-      "image/png": "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",
-      "text/plain": [
-       "<Figure size 640x480 with 2 Axes>"
-      ]
-     },
-     "metadata": {},
-     "output_type": "display_data"
-    }
-   ],
-   "source": [
-    "corr_matrix = non_winning_trades[[\"confidence\", \"win_probability\"]].corr().round(2)\n",
-    "sns.heatmap(corr_matrix, annot=True)\n",
-    "plt.show()"
-   ]
-  },
-  {
-   "cell_type": "code",
-   "execution_count": null,
-   "metadata": {},
-   "outputs": [],
-   "source": [
-    "del tools\n",
-    "del tools_inc\n",
-    "del tools_non_error\n",
-    "import gc\n",
-    "gc.collect()"
-   ]
-  }
- ],
- "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": "3.12.3"
-  }
- },
- "nbformat": 4,
- "nbformat_minor": 2
-}